For information on obtaining additional copies, reprinting or translating articles, and all other correspondence,

please contact:
Email: InfosyslabsBriefngs@infosys.com
Infosys Limited, 2012
Infosys acknowledges the proprietary rights of the trademarks and product names of the other
companies mentioned in this issue of Infosys Labs Briefngs. The information provided in this
document is intended for the sole use of the recipient and for educational purposes only. Infosys
makes no express or implied warranties relating to the information contained in this document or to
any derived results obtained by the recipient from the use of the information in the document. Infosys
further does not guarantee the sequence, timeliness, accuracy or completeness of the information and
will not be liable in any way to the recipient for any delays, inaccuracies, errors in, or omissions of,
any of the information or in the transmission thereof, or for any damages arising there from. Opinions
and forecasts constitute our judgment at the time of release and are subject to change without notice.
This document does not contain information provided to us in confdence by our clients.
VOL 10 NO 2
2012
SySTEmS
INTEgraTION
I
n
f
o
s
y
s

L
a
b
s

B
r
i
e
f
i
n
g
s
S
Y
S
T
E
M
S

I
N
T
E
G
R
A
T
I
O
N
V
O
L

1
0

N
O

2

2
0
1
2
Subu Goparaju
Senior Vice President
and Head of Infosys Labs
At Infosys Labs, we constantly look for opportunities to leverage
technology while creating and implementing innovative business
solutions for our clients. As part of this quest, we develop engineer-
ing methodologies that help Infosys implement these solutions right,
frst time and every time.
Infosys Labs Briefings
ANIL PRASAD KURNOOL is a Senior Technology Architect with ECS vertical of
Infosys. He can be reached at AnilPrasad_Kurnool@infosys.com.
ANIRBAN GHOSAL is a Principal Technology Architect with the RCL business unit
of Infosys. He can be contacted at anirbang@infosys.com.
BALAJI SUBBARAMAN was a Technical Architect with Infosys Cloud Computing
Practice.
GANESH JAYARAMAN is Senior Consultant with the Communication domain and
process consulting practice. He can be reached at Ganesh_jayaraman01@infosys.com.
KARTHICRAJA GOPALAKRISHNAN is a Senior Technology Architect with
the Financial Service and Insurance Business Unit of Infosys. He can be contacted at
Karthicraja_G@infosys.com.
LADISLAV GATA was a Senior Technology Architect with the Consulting & Systems
Integration Practice for Retail Business Unit of Infosys.
N. VIJAYKUMAR is a Principal Technology Architect with Infosys Cloud Computing
Practice. He has around 18 years of experience in the IT Infrastructure domain, both in
consulting and implementation.
NARAYANAN CHATHANUR is a Senior Technology Architect with the FSI business
Infosys. He can be contacted at narayanan_cs@infosys.com.
PRAKASH RAJBHOJ is a Principal Technology Architect with the rcl business unit of
infosys. He can be reached at prakashr@infosys.com.
RAMKUMAR DARGHA is a Principal Technology Architect with Cloud Computing
practice. He can be contacted at Ramkumar_Dargha@infosys.com
ROHIT CHRISTOPHER Rohit Christopher is a Senior Technology Architect with Cloud
Computing practice. He can be contacted at rohit_christopher@infosys.com.
SAURABH JOHRI a Senior Manager in the Corporate Planning unit of Infosys.
SAURABH KUMAR MISHRA is a Technology Architect with Infosys cloud
computing practice. He is currently focusing on Big Data Solutions and can be contacted at
saurabhkumar_mishra@infosys.com.
SENTHIL KUMAR PR is a Senior Technology Architect with the Manufacturing
vertical of Infosys. He can be reached at Senthilkumar_pr@infosys.com.
SUCHIT BHATWADEKAR is a Principal Technology Architect with thethe
Consulting and Systems Integration Practice of Finacle at Infosys. He can be contacted at
suchit_bhatwadekar@infosys.com.
The earlier half of IT evolution was primarily focused on automating
time consuming and labor intensive tasks. Later, systems were built to
automate all and sundry organizational processes. as information systems
made workforce productive, many state-of-the-art systems were built in
silos targeting a particular problem. These systems were built using varied
languages and on varied environments. a need was then felt to integrate
these disparate systems to build a system that has best of all worlds
functionality. Systems Integration (SI) therefore emerged as a field that
brought together disparate systems to operate as one single system.
Today the role of systems integrators is becoming increasingly important
as more and more systems are being developed and such newly built
systems are to be in sync with the existing systems. The job of the systems
integrators is therefore that of developing interfaces to new and existing
systems such that they glue together as one system.
modernization of existing platforms has in a way become the need of the
hour and no sane enterprise can go about engaging in a modernization
initiative minus a clearly defined methodology. One cannot do away with
existing legacy systems, not only because of the huge replacement costs that
enterprises will have to incur, but also because of the inherent advantages
that they have in terms of being storehouses of enterprise processes and
data. Therefore, a better and cleverer way of embracing modernity is by
integrating legacy systems with todays enterprise applications. Enterprises
now are therefore not to shed their legacy systems but modernize them
leveraging SI expertise.
SI has traditionally been looked upon as a deployment and maintenance
function. However, todays SI has the capability to partner enterprises in
their business transformation journey. In this collection we have papers that
shed light on how systems integrators can play a key role in modernizing
legacy and helping transform organizations into smarter organizations.
your feedback is important to us. as usual, do write to us with your
suggestions on how we can further improve your reading experience.
yogesh Dandawate
Deputy Editor
yogesh_dandawate@infosys.com
Authors featured in this issue
Infosys Labs Briefings
Advisory Board
Anindya Sircar PhD
Associate Vice President &
Head - IP Cell
Gaurav Rastogi
Vice President,
Head - Learning Services
Kochikar V P PhD
Associate Vice President,
Education & Research Unit
Raj Joshi
Managing Director,
Infosys Consulting Inc.
Ranganath M
Vice President &
Chief Risk Officer
Simon Towers PhD
Associate Vice President and
Head - Center for Innovation for
Tommorows Enterprise,
Infosys Labs
Subu Goparaju
Senior Vice President &
Head - Infosys Labs
modernize Legacy with
Systems Integrators
Infosys Labs Briefings

3
7
15
21
29
35
45
53
59
VOL 10 NO 2
2012
Discussion: Why Enterprise Customers Require Integration with Legacy System?
By Anil Prasad Kurnool and Senthilkumar P.R
Seamless integration of enterprise applications is a must in todays competitive world.
With more and more devices and platforms coming into existence the need for
integration of enterprise applications has become stronger and all pervasive.
Trend: Complex Events Processing A New Trend
By Karthicraja Gopalakrishnan and Ladislav Gata
Complex Event Processing (CEP) is emerging in an era when service oriented
architecture (SOA) and business process management (BPM) are reaching their
maturity. In this paper the authors analyze how CEP can provide significant
benefits to organizations.
Viewpoint: How Systems Integrators Can Help Build Smarter Organization?
By Saurabh Johri and Ganesh Jayaraman
Enterprises are looking upon System Integrators skills to aid them in
their strategic journey and become smarter organizations.
Analysis: Effectiveness of EA Frameworks in Achieving BITA:
An Analysis Based on SAMM
By Ramkumar Dargha
Business IT Alignment (BITA) is one of the main goals of all enterprise architecture
(EA) initiatives. The author draws from his practical experience in discussing
the applicability of SAMM model to the popular EA frameworks like TOGAF,
ZACHMAN and analyzing the effectiveness of the support provided by these
frameworks in achieving BITA.
Approach: Proactive Performance Engineering for E-commerce Modernization
By Saurabh kumar Mishra, Balaji Subbaraman and Rohit Christopher
Modernization of existing platforms is the need of the hour and enterprises are
engaging in a modernization initiative without a clearly defined methodology.
The authors propose a simple approach that would ensure the success of the
modernization program.
Model: How Systems Integration Helps Achieve Progressive Business Transformations?
By Anirban Ghosal and Narayanan Chathanur
Business transformation is aimed at producing enterprise level performance
improvement through radical changes in the business and IT systems. In this paper,
the authors present a view on how this complex process of progressive transformation
can be achieved through simple steps of aligning the business vision with IT vision.
Framework: Applications Integration with ESB
By Prakash Rajbhoj and Narayanan Chathanur
All types of enterprises whether small or large always have the need for EAI.
Traditional EAI is expensive and has its own incumbent problems. In this paper the
authors show how enterprises can create an integration platform for their IT systems
by using the pillars of EAI and SOA and solve their business problems.
Insight: Solution Architecture in Large Transformation Programs
By Suchit Bhatwadekar
Solution architecture is a framework to develop end-to-end view of change that fully
caters to the business needs, ensures technical compliance within budget and has a
maintainable schedule. In a transformation program, success of any solution is heavily
dependent on the acceptance of the solution by the relevant stakeholders opines the
author.
Practioners Perspective: Role of Systems Integrators in the Co-creation Process
By N.Vijaykumar
Co-creation is a budding discipline of the current wave of innovation. Collaborative
creativity focusing on people and aiming to enhance the value provided to consumers
is the focus of co-creation. In this paper the author explains how systems integrators
can help organizations realize the co-creation dream.
Index

In a transformation program, success of any
solution is heavily dependent on the acceptance
of the solution by the concerned stakeholders.
Suchit Bhatwadekar
Principal Technology Architect
Finacle Consulting and
System Integration Practice,
Infosys Limitied.

Successful systems integration has enabled organizations
improve operational efficiency with better reaction time,
information accuracy and integrity.
Karthicraja Gopalakrishnan
Senior Technology Architect
Financial Service and Insurance
Consulting & System Integration Practice,
Infosys Limited.
3
VOL 10 NO 2
2012
Why Enterprise Customers Require
Integration with Legacy System?
Enterprises can ill afford to shed
legacy systems completely
By Anil Prasad Kurnool and Senthilkumar P.R.
S
eaml es s i nt egr at i on of ent er pr i s e
applications is a must in todays competitive
world. With more and more devices and
platforms carrying data processing muscle
- the most common example being mobile
phones, the need for integration of enterprise
applications has become stronger and all
pervasive.
For the enterpri ses to achi eve the
elusive agile IT infrastructure all the systems
need to be integrated and data flow must
happen without hindrance. Legacy never
dies! goes an old adage. In an industry of
faster hardware (Moores Law), fast moving
data (IPv6 networks) and better applications
development methodologies (OOPS, 4GL)
legacy system still strives and holds its own
fort, making it in most of the cases the back
bone of enterprise IT.
CIOs today are caught in the perpetual
dichotomy of ensuring that their IT supports
the fast changing business dynamics even
while retaining legacy systems as part of the
endeavor.
IMPORTANCE OF AND NEED FOR LEGACY
From the perspective of stability, CIOs hark on
the benefts that legacy systems have to offer in
comparison to the current systems and hence
legacy systems are the most preferred choices to
host core applications of enterprises. Robustness
is another key feature that makes legacy systems
extremely reliable and promising [1].
In addition, one cannot discount the fact
that an enterprises historic information that
is stored deep within legacy systems plays an
important role while collaborating with external
partners and stakeholders.
From a pecuniary perspective too,
legacy systems score heavily in comparison
to niche technologies primarily because of
three reasons easy availability of talent
pool to maintain the legacy system; cost
effectiveness in operations and the need for
fresh investments in replacement technologies.
Legacy system can also be considered
as a source of competitive advantage to the
enterprise as the logic residing in it makes it
diffcult for the competitor to emulate.
Infosys Labs Briefings
4
Some popular reasons that make the legacy
systems the most preferred choice are mentioned
below.
Cost of redesi gni ng/repl aci ng a
legacy system is substantial due to the
complexity of the functionality. The
decision of the replacement or redesign
depends on the cost-beneft analysis.
Di smant l i ng of t he syst em mi ght
affect a business function, if no proper
equivalent alternate arrangements are
considered. This reinforces the above
mentioned reason that complexity in
building a new application on lines of
the functionality present in legacy is far
from welcoming.
Lack of suffcient technical knowledge
of the current generation development
community to understand the legacy
system and absence of documentation
of the algorithms used to develop the
applications forms a formidable challenge
when one wants to replace legacy [2].
CFOs lack of interest in exploring new
IT ventures, especially when the legacy
systems serve the current purpose.
Also failure witnessed in migration and
modernization initiatives is a stumbling
block that fuels the fear of a CFO from
making any new investments in newer
technologies [3].
Despite the hue and cry to migrate
to more sharper and efficient technologies,
legacy systems are going to stay for the reasons
mentioned above. So it would make sense for
enterprises to integrate the advantages that
legacy systems have to offer to realize the much
needed business agility.
NEED FOR INTEGRATION WITH LEGACY
To stay ahead of the competition companies
must adopt cost effective IT solutions and be
more agile in their business. The pace at which
technology evolves and the resulting changes
have become a constant challenge to the
enterprise IT systems. It is not cost effective to
replace or redesign the applications whenever
there is a change in technology.
Many enterprises would have a set of
applications created for each of their business
lines. Over a period of time, enhancements
and enrichments made to these systems might
have pushed them towards becoming islands of
information or silos. Such silos can be reckoned as
legacy systems. And today, most new initiatives
that enterprises engage in may need data from
different silos. It is here that the business value of
data hidden in the legacy systems can be leveraged
in new business processes and operations by
providing effcient ways to integrate.
Legacy systems do not have their business
logic/functionalities in a readily usable form.
Till their core functionality is made available
in a reusable form, their features might not be
providing the needed value to the business. This
reusability aspect makes the integration of the
legacy systems with other applications a highly
prudent investment for the business.
New business regulations might require
signifcant changes and customization to the IT
infrastructure for audits. Data residing in legacy
system can be harnessed by integrating the
legacy system with other applications thereby
bringing in the much needed transparency in
audit exercises.
This integration approach reduces the go-
to-market time for any new business initiative.
5
Other needs that can prompt integration
with legacy would be a new business process
optimization initiative, or even a lack of a
complete new commercial off-the-shelf (COTS)
application that can support entire business
needs of an enterprise.
It might be faster to integrate and
assimilate a legacy system than to replace it. As
an additional beneft, even while considering
the replacement of legacy due to shelf life issues,
it is benefcial to go in with an integrated legacy
system approach.
Once integrated with the perspective
of retirement, any legacy application can be
moved from the core of the IT to be a deemed
as component of IT. This shift would enable
the CIO to segregate the coarse grained
functionalities hosted in legacy system along
with their auxiliary fne grained functionalities
into buckets and then start looking out for
alternatives to these buckets.
Once an alternative is found, then the
bucket from legacy can be retired and the
alternate can be used. This co-existence of
legacy and its alternate may seem to be a long-
drawn retirement plan, but the integration
mitigates most of the risks involved in big-bang
approach, as the hot-swap is ready, in case of
any eventuality and the change needs to be
rolled back as the legacy is still integrated with
the current IT landscape.
PATTERNS OF INTEGRATION FOR
LEGACY SYSTEMS
Todays integration technologies have different
and innumerable techniques for integrating
with the legacy applications. These can be
broadly classifed as --
Screen Scraping Technique: This technique is
typically used for terminal based legacy system
where the data on a screen is read through an
auxiliary port. The data is fed to a form on the
screen too by writing data to the terminals
memory. This technique mimics a human
interaction with a terminal (also called dumb
-terminal). Even in cases where the existing
application might not be having any known
application programming interface (API) this
technique can be used. This technique can be
used even when the source code of the system is
too complicated to understand, just by writing
directly into the memory of the application.
In a nutshell this is a non-intrusive technique.
Message-based Interactions: IBM s MQ
Series pioneered this technique of message-
based interaction. The product suite or the
middleware, also called Message Oriented
Middleware, can be hosted on a very wide
range of hardware and heterogeneous operating
systems, ranging from Mainframes to AS/400 to
WinTel. Applications considered for integration
need to only communicate with the local agent
on the same box that can provide a message
to the application in the form of subscription
or in a request-response fashion. Based on the
use case, this technique can be either intrusive
(making changes to application to accept
message based instructions) or non-intrusive
(where the application implicitly understands
the contents and context of message.) This
technique is currently in vogue in most legacy
system based enterprises.
Transaction Wrapping: If the legacy application
uses any of the transaction systems like CICS,
these transactions are wrapped by agents hosted
on the legacy applications box itself. Starting
with ver 3.0, CICS Transaction Server from IBM
began providing wrapping services out-of-box,
in form of web-services [4].
6
BENEFITS OF INTEGRATING LEGACY
SYSTEMS
Of the many options and routes taken by
enterprises for IT agility business process
management (BPM) and services are the
most preferred options. BPM solutions help
enterprises adapt to the fast changing market
conditions and business opportunities quickly
by providing the required tools to the business
users to change the business processes with
relative ease.
The crucial differentiator of a BPM
solution from other IT solutions is that BPM
can be used to create new value by compositing
new business services out of the existing
business systems without impacting the
existing landscape. In this context the agility
derived from BPM-based services includes
legacy systems services without changes
to them. Preserving and extending legacy
systems through BPM-based solution allows
the enterprises to retain their key systems
and capitalize on their strengths. This also
mitigates the potential risk associated with
replacing mission critical applications hosted
on legacy systems. By integrating the legacy
systems business processes can leverage the
data that is residing deep inside them. Through
integration, a layer of abstraction is provided
to the landscape, which can be used and reused
by services and BPM without impacting the
underlying implementation in the legacy
system or in any other system.
CONCLUSION
Modern enterprises need dynamic and agile
systems to take care of their business demands.
However, enterprise IT landscapes are dotted
with the presence of legacy applications. A
typical legacy system is too expensive and risky
to be done away with.
Robustness, reliability and security
offered by legacy systems make them the
most preferred choice to host core business
functionalities. Usually they earn the patronage
of the CxO community from their cost effective
operations, abundant talent pool and historic
data reasons. However, in any initiative to
automate or SOA-enable an enterprise to cater
to the growing business demands, legacy
systems become the proverbial last mile to
achieve the needed IT agility.
Legacy systems are here to stay and
enterprises need to explore the options of
inclusive agility by integrating them with
other applications instead of replacing them.
Once integrated, the Legacy systems would
prove to be a cost effective way of providing
the functionality needed by business in a very
robust, secure and a reliable way.
REFERENCES
1. Soltis, F. (September 2002), Delivering
Mainframe Advantages to Your Business.
In The Power of Mainframe Computing.
Avai l abl e at http: //www-07. i bm.
com/servers/eserver/au/i seri es/
downloads/i890_fnal_whitepaper.pdf.
2. Lack of documentation (Definition).
Available at http://www.cc.gatech.edu/
morale/local/morph_glossary.html.
3. Bergey J., Northorp L., Smith D. (October
1997), Enterprise Framework for the
Disciplined Evolution of the Legacy
Systems. Available at http://www.sei.
cmu.edu/reports/97tr007.pdf
4. Rayns, C., Burgess, G. ,Cooper, P. et.al
(January 2010), Application Development
for CICS Web Services IBM Redbook,
Chapter 2, ISBN 0738433853. Available
at http://www.redbooks.ibm.com/
abstracts/sg247126.html.
7
VOL 10 NO 2
2012
CEP is fast emerging as a major breakthrough
for mainstream adoption in enterprises than
merely being an extension to concepts
like SOA and BPM
Complex Events Processing
A New Trend
By Karthicraja Gopalakrishnan and Ladislav Gata
Infosys Labs Briefings
F
or many years, once of the key success
factor for business process automation
has been enterprise application integration
(EAI), which enabled enterprises to make their
applications talk to one another though they
were not originally built for such dialogues.
Successful systems integration has enabled
organizations improve operational effciency
with better reaction time, information accuracy
and integrity. Complex Event Processing
(CEP) elevates these benefts to the next level
by enabling organizations link interrelated
business events arising out of the enterprise
systems and develop powerful real-time
business intelligence and cognitive business
processes.
CEP creates such benefts by enabling a
very powerful event defnition and detection
capability in the enterprise system integration
landscape. CEP as a market solution on the
one hand provides mechanism for the user to
defne event patterns and deploy, and on the
other hand provides a powerful engine that
can recognize event patterns from the cloud
of enterprise system events and trigger an
automated response. Due to its event processing
power, CEP systems can be made to learn and
apply knowledge automatically and made
cognitive and capable of changing preferences
without enforcing a programmed path. In this
paper, we analyze how CEP can contribute
signifcantly to the competitive advantage of
organizations.
CEP is emerging in an era when service
oriented architecture (SOA) and business
process management (BPM) are reaching
their maturity. As SOA and BPM have been
signifcantly affecting the system landscape in
recent years, the impact of CEP on enterprise
architecture very much depends on how it will
8
marry with these major trends. CEP is joining
SOA and BPM on mission to make enterprise
systems more open to frequent changes
implied by increasing dynamics of the market.
CEP brings an architecturally strong concept
that enables people define more complex
and more dynamic business processes, but
more importantly it can make systems learn
how to optimize business processes without
human intervention. Although CEP has set
off on journey towards mainstream enterprise
technologies mainly by flling the gaps in BPM
and SOA, in the long term it can overrun these
concepts completely and might lay down the
foundation for a new generation of enterprise
systems powered by central information
backbone and central intelligence.
WHAT IS COMPLEX EVENTS PROCESSING?
Enterprises have tried to surface business
significant events from their mainstream
systems for many years now. They were in
the form of specialized handling of conditions
like raising an alert when the inventory drops
to a certain level, or executing an alternative
operation when business exceptions occur or
automatically escalate priority if a particular
business process has not met its service
level agreement, etc. However, these events
were implemented with in an application or
a program module and traditionally had a
very limited scope. Secondly, the concept of
analyzing important enterprise events did
exist in the past in the form of management
information reports; however, they did not
cater to producing real time automatic response
or understanding interrelated complex events.
CEP bridges this gap and takes event processing
to the next level in the enterprise. We defne CEP
as an enterprise system capability to analyze
and interpret the complex series of interrelated
events arising out of the enterprise systems with
an aim to recognize business signifcant events
in real-time. The recognition of an event can be
used for different types of actions. It can alert
humans if the event needs a user attention; it
can trigger automated action in various systems;
or it can distribute information provided by the
event across enterprise.
Some organizations have successfully
implemented CEP for many years in the felds
of algorithmic trading, traffic management
fraud detection, anti-terrorism and in expert
military systems. They have significantly
improved human effciency in understanding
and processing business significant events
from the endless array of information pipeline.
However, the benefits of the CEP have not
reached the enterprise systems.
WHY IS CEP A POWERFUL CONCEPT?
CEP can help enterprise IT and business in
many ways. A CEP application can proactively
recognize a business significant event and
automatically trigger a corrective action or
can surface the complex and behind the scene
details of a seemingly known event to enable
corrective work on the real problems or develop
a central intelligence hub in a non-invasive
manner to control an existing business process
or build a dynamic business process that is
defned by states than steps. To understand
how CEP delivers such power and benefts it
is essential to understand the anatomy of a
CEP system.
CEP is an event processing system.
In the center of this powerful system lies a
data processing engine capable of handling
huge data correlation in memory. The data
correlation engine builds an event pattern
recognition component to form the core of
CEP. CEP solution development is about
9
designing event patterns and is developed
by business analysts using the event pattern
modeler. Figure 1 depicts the key components
involved in a CEP system.
This architecture of CEP provides the
power and effectiveness in a number of ways;
CEP architecture is event based hence does not
manifest its presence in other applications in the
enterprise landscape and can easily plug into an
existing infrastructure to consume the already
exposed messages. Its capability to process
huge data in-memory under million operations
per second enables real-time analysis of current
and historical data. The adapters and agents
allow consuming data from disparate sources
and fnally the action manager allows triggering
compensating operations orchestrated from
outside the existing enterprise applications.
CEP is extremely powerful because of its huge
power to analyze, predict, execute and learn
from a huge pile of information from disparate
sources in real-time and more importantly, in a
non-invasive manner.
HARNESSING THE POWER OF CEP
Though originally conceptualized as an advanced
monitoring solution, CEP today is expanding
in different areas of enterprise systems by
providing real-time intelligence it brings in
the next level of business process visibility inV
order to enable on-time and accurate business
decisions; by enabling unstructured business
process defnition; by making processes more
complex, dynamic and fexible and by bringing
in aspects of expert systems to enterprise
environment and laying the foundation of future
cognitive enterprise systems.
Real-time Intelligence
Business intelligence has been a batch processing
functionality for enterprises in the past. They
produced daily, weekly and monthly reports
to present a historical view of the business
effciency. CEP takes this to the next level by
providing real-time business intelligence which
means corrective actions and optimizations can
be made instantaneously.
Another important beneft of CEP is that
it can subscribe to events from disparate sources
it can be a network event, or storage device
event or events from business systems. CEP
can consume no matter where the event was
generated. Detection of semantically equivalent
fragments from such varied sources has never
been possible before and CEP has enabled it.
That means business intelligence can be robust
and extensive.
Unstructured Business Processes
For many years BPM solution strategy assumed
that the business processes are sequential and
predefned in certain structure which we call
procedural BPM. Over years of implementation
a better understanding has emerged from
experience that the business process are
unstructured to the tune of about 60-70% and
Figure 1: Key Components in a CEP System
Source: Infosys Research
Event Pattern Modeller
(Design-time)
Event Pattern Recognition Engine
Message Bus
Database
Rule Engine
Action
Manager
Agents/
Adapters
10
not supported suffciently by the current BPM
products in the market [2].
As an example, it is not hard to relate
in your enterprise that on many occasions the
user of the business processes wants to change
the way the process is executed. She may want
to skip one of the steps due to a newly arisen
situation or perform the 4th step before the 3rd
step or perform 3rd and 4th step in parallel or
breakdown the 3rd step into two and perform
in the order of 3a, 4, 3b. The point is the
sequence of the business process can change
very frequently.
The current BPM products in the market
do not cater to this fundamental change in
the way workfows are defned and are hence
fairly ineffcient in delivering to the business
requirements. In situations like these CEP
brings in a major advantage. It can be used for
developing unstructured business processes
by putting the user focus on to advancing
the business process states than advancing
through the steps. This is a major shift in the
way workflows are defined, modeled and
implemented today. One of the real world
examples of such process is mortgages and
loans that have a very long and complex
lifecycle with many variations of loans and
mortgage products. It is too diffcult to model
in a procedural way.
Cognitive Systems
Artifcial intelligence has not been used much
in enterprise business processes partly due to
the complexity involved in developing them
and partly because of the capacity required in
its execution. The characteristics of CEP such as
loosely coupled event based architecture, huge
in-memory processing power and the capability
to recognize event patterns can form a foundation
for implementing expert systems in enterprises.
A CEP system can be designed such
that it can analyze the user behavior, preserve
knowledge and enable automated reasoning
when there is a high probability of user
repeating the same action. Self optimizing
business process can be made a reality with the
introduction of artifcial intelligence into the
enterprise applications.
There is no doubt that CEPs potential of
providing added business value is more than
significant and CEP will become a mainstream
adoption for enterprises. However the CEP
j ourney into the mainstream will not be
easy, as the concept is not just filling a gap,
but at certain level also interferes with other
ambitious and more established concepts like
SOA and BPM.
CEP AND FUTURE SYSTEM INTEGRATION
Why should CEP be discussed in context of
system integration? To give an answer we have
to understand the role of integration in modern
enterprise architecture and the direction where
it is heading to. Integration platforms evolution
undoubtedly converges to enterprise service bus
(ESB) concept. ESB driven by SOA concept has
changed the traditional enterprise application
integration (EAI) middleware from a plumbing
infrastructure to a standardized enterprise
backbone exposing business functions. The
evolution of enterprise IT is coming close to
the end of one era where business process
is exclusively owned by applications. ESBs
today have capability to host process engines
and containers for mainstream programming
languages. The standardized environment and
strong component model of ESB will attract
more and more business logic to be hosted in
this ecosystem. In the more distant future this
trend may lead to architecture of centralized
backbone with central intelligence.
11
We can fnd a direct analogy in nature
and compare ESB to nervous system of living
creatures. The central nervous system (CNS)
comprising brain and spinal cord is controlling
all body parts through peripheral nervous
system (PNS). The enterprise spinal cord is
the standardized message broker contained by
ESB, the brain is the integrated process engine
and the body parts are various enterprise
applications [Fig. 2]. The PNS is the integration
capability provided by different adapters and
bindings which connect applications to the
central message broker.
CEP will be a critical aspect of making
the vision of central intelligent hub a reality as
this concept will need a more sophisticated brain
than a process engine with procedural logic. To
enable continue business process optimization
the enterprise central intelligence will need to
employ rule-based reasoning engine or other
artifcial intelligence mechanisms in order to
correlate and react to millions of signals coming
from various applications. This sort of capability,
typical for expert systems, is what CEP lays the
foundation of in enterprise environment. Even
though CEP is used for specifc tasks in various
industries for decades, it is system integration
evolution which brings CEP to the mainstream.
This mainstream can be seen as enterprise
architecture with central information backbone
and central intelligence evolved from CEP
enabled ESB, an enterprise with CNS.
CEPS IMPACT ON SOA AND BPM
SOA has affected enterprise architecture
dramati cal l y wi th the evol uti on of ESB
technology. It has pushed boundaries of system
integration and has changed the way how
applications are built. BPM platforms originally
developed as standalone tools are becoming
part of the SOA reference architecture offering
service orchestration. SOA and BPM hand in hand
are making an attempt to build new enterprise
architecture paradigm viz., publish your business
functions on the bus and model your business
process in BPM tool. The main driver is fexibility,
easy creation and change in business process
reusing services. In order to achieve that, ESB
products are becoming tightly integrated with
BPM platforms or directly host BPM engines.
CEP is a new arrival to this family with clear
ambition to make the business process even more
fexible and dynamic. So, the biggest question
that remains to be answered is how the marriage
between CEP, SOA and BPM is going to end.
SOA i s st i l l percei ved mai nl y as
architecture style where one application can
provide business functions consumed by other
applications. In order to manage the intra
application dependencies SOA employs ESB
as broker between consumer and provider
applications. This SOA based view of ESB
mediating and exposing business functions
for remote invocation has kept out of focus the
fact that ESB is messaging based with ability to
send and receive events. For that reason CEP as
event based infrastructure is fully compatible
with ESB concept and in the next few years CEP
is not going to bring any fundamental change
in ESB evolution.
Enterprise Central Intelligence
Internet
Enterprise
Monitoring & Management
Enterprise Information Backbone
Mainframes & Legacy Applications Data Centres
Figure 2: Central Intelligence of Enterprise
Source: Infosys Research
12
ESB provides abstrtaction of different
appl i cat i ons f unct i ons and event s i n
standardized environment. CEP and BPM
technologies can be hosted in this environment
in order to orchestrate those events and
functions into business logic. BPM enables to
build procedural based logic, CEP is typically
rule based. CEP as paradigm has advantage as
it can model any real world scenario whereas
procedural modeling cannot. Apparently, this
may cause an architectural confict since in many
situations both the concepts are applicable. In
the long run the stronger CEP concept will very
likely overtake the mainstream and accompanied
with other expert system disciplines will become
a key to the next generation of intelligent
enterprise systems. Until this happens CEP and
procedural modeling will coexist in ESB ecosystem
[Fig. 3].
In the next few years, CEP will be
augment more matured BPM platforms making
business processes more dynamic. From pure
system integration perspective, CEP engine as
integral part of ESB will be used for scenarios
where rule based routing and correlation
of events is needed. CEP will also help to
understand and use better the eventing aspect
of ESB which has been often overlooked.
Business activity monitoring as the most visible
CEP discipline today will very likely also enter
the ESB ecosystem in order to get closer to the
source of business events.
MATURITY OF CONTEMPORARY
CEP TECHNOLOGIES
CEP products are dynamically developing
and they are catching up with the evolution of
SOA technologies. Most middleware vendors
are enriching their SOA product stack with
a CEP platform which is often referred to as
event-driven architecture (EDA). The term
EDA is not used consistently by all vendors as
in some cases it can indicate ESB capability to
provide event based interaction without CEP
features offering.
The important characteristic of CEP
technologies is diversity. The diversity is
caused by two key factors: (a) variability of CEP
applications in enterprise environment; and (b)
lack of developed and adopted CEP standards.
The vast majority of CEP products are platforms
that can be used standalone or in conjunction
with other systems including middleware.
There are also emerging ESB products and BPM
suites that may offer integrated CEP features.
In general the standalone CEP platforms
are more mature because of longer history.
Another common symptom i s that non-
functional capabilities like development,
monitoring and management tools or design
time governance are not in the spotlight yet
and we can expect dynamic growth of these
aspects with the growth of CEP deployments
and increased overall CEP acceptance by
mainstream enterprise technology market.
APPROACHING A CEP IMPLEMENTATION
CEP implementation is for many companies a
new area which is a major risk that needs to be
acknowledged in the frst place and all necessary
Internet
BPM
Engine
CEP
Engine
Message Broker
Mainframes & Legacy Applications Data Centres
ESB
Figure 3: Coexistence of CEP and BPM in ESB Ecosystem
Source: Infosys Research
13
project management mechanisms should be
employed to mitigate that risk. Apart from that
the main challenges of CEP adoption are:
Identifying the business need for CEP
Understanding the technical nature of
the business need
Selecting and understanding the product
Defning CEP development methodology
Implementing CEP in the intended
fashion.
How does a company then recognize the
need for a CEP? In an earlier section we have
elaborated different areas of CEP applications.
Business requirements addressing some of
those areas should be the frst indication of a
possible CEP need. Subsequently, the particular
business scenario has to be scrutinized in order
to correctly understand its technical nature.
Just precisely derived technical requirements
can justify whether CEP platform would bring
an added value.
In the CEP product selection process
the technical requirements play a key role as the
CEP platform is a sort of software infrastructure
without out-of-the-box business functionality,
therefore the evaluation is focused purely on the
technical aspects. The selection process is more
challenging, as for instance, unlike application
servers or ESB products, CEP platforms are
not standardized. By evaluating an application
server in the current market the license cost and
level of support might be more important than
actual capabilities. Standardization has brought
many vendors so close in terms of features that
product selection does not mean a signifcant
risk. This is not the case for CEP today. Detailed
technical requirement specifcation and proper
understanding of the product can make a huge
difference in project execution.
To get the right confdence it is critical
to work closely with vendors in the product
selection process. There are no standard
methodologies for CEP development, hence
getting close to the product before it gets
selected will help one to identify any additional
challenges and risks of delivery and maintenance
process.
CEP implementation shows symptoms of
high risk projects, hence solution prototyping
and incremental delivery process is preferable
and can help fnalize the project in intended
fashion.
CONCLUSION
The CEP as a mainstream is at the beginning, but
should not be ignored. To leverage the full power
of CEP may require radical transformation of
the entire system integration landscape. CEP
function is to capture events and trigger actions
in real-time. In this aspect SOA maturity can
be a critical factor since in an environment
where business events and business functions
are available in standardized ESB ecosystem,
future CEP implementation can be achieved
with lower risk and higher added value. Thus
even if a CEP project is not on your short-term
roadmap, including the CEP readiness in the IT
strategy is the right step to do today.
Although CEP implementation may
be seen as high on investment risk in terms
of project realization, the possible benefits
it can yield may be significant. The added
busi ness val ue may bri ng an i mportant
competitive advantage and major breakthrough
in companys business. Bringing fexibility of
business process to a new level and ensuring
that decision-making information is available
in real-time can open new opportunities of
business innovations and major improvements
in operational excellence.
14
4. Vollmer, K. (2009), Forrester Wave
Comprehensive Integration Solutions
Q4 2010 for Application Development
& Delivery Professionals. Available at
http://www.oracle.com/us/corporate/
analystreports/infrastructure/forrester-
cis-wave-2010-189779.pdf.
5. Gualtieri, M., Rymer, J. (2009), The
Forrester Wave- Complex Event Processing
(CEP). Available at http://www.forrester.
com/rb/Research/wave%26trade%3B_
compl ex_event _pr oces s i ng_cep_
platforms%2C_q3/q/id/48084/t/2.
6. Event-Driven SOA: A Better Way to SOA
(2011). Available at http://www.tibco.
com/multimedia/wp-event-driven-
soa_tcm8-803.pdf.
REFERENCES
1. Luckham, D. (2001), The Power of Events:
An Introduction to Complex Event
Processing in Distributed Enterprise
Systems. Addison-Wesley Longman
Publishing Co., Inc., Boston, MA, USA.
2. Luckham, D. (2001), A Short History of
Complex Event Processing. Available
at http: //complexevents. com/wp-
content/uploads/2008/02/1-a-short-
history-of-cep-part-1.pdf.
3. Mishra, R. (2010), Unstructured business
processes - creating differential BPM
strategy is the wise thing to do by Rakesh
Mishra. Available at http: //www.
infosysblogs.com/bpm-eai/2010/01/
unstructured_business_processe.html.
15
VOL 10 NO 2
2012
Harness the potential of your systems
integrator to build a smarter organization
How Systems Integrators
Can Help Build Smarter Organization?
By Saurabh Johri and Ganesh Jayaraman
A
s organizations are increasingly becoming
global their operations are expanding
across geographies and new products and
services are being introduced at a rapid pace.
Consequently, enterprise complexity in terms
of business processes and underlying IT
systems has also exploded. The ever increasing
complexity over time reduces effciency, speed
and accuracy of operations.
Organizational complexity has cascading
impact in multiple areas of business functions.
For example, an inefficient customer service
management system can adversely impact
customer experience and future revenues. The
organizations inability to accurately view supply
chain costs will cause reduced proft margins.
Whi l e organi zat i ons scal e up, I T
landscapes typically evolve over time as
disparate subsystems in silos. Such systems
with limited interoperability offer limited
view of organizations state and the entire
business picture. Furthermore the total cost of
ownership increases due to high maintenance
costs. According to a Gartner study, typically
67% of IT spends is directed towards keeping
the lights on operations which leave little for
new transformational projects [1]. With new
technology disruptions like cloud computing
and mobility gaining traction, an ineffcient IT
organization will impact frm competitiveness.
System integrators (SIs) have vast
expertise in bringing disparate systems together.
System integration skills can aid organizations
in the strategic journey to become smarter
organizations.
FACTORS LEADING TO COMPLEXITY
A close look at a typical IT landscape of a
large organization will reveal mishmash of
legacy systems, commercial off-the-shelf
(COTS) ERP and CRM applications and a
host of custom applications. Decentralization
of the IT function and its alignment to a
geography or line of business (LoB) leads to
multiple instances of the same application
and duplication of business functionality.
Enforcing a consistent IT governance policy
and application landscape is a key challenge
in the face of rapid changes in the business and
technology environment.
Infosys Labs Briefings
16
Business Drivers: As markets are getting
liberalized, companies are increasing their
global presence extensively.
Mergers and acquisitions are happening
on a continual basis and integrating the IT systems
of two different frms is a complex task involving
the reconciling of duplicate data and rationalizing
the different storage systems and applications.
Organizations exploring the organic
route to growth introduce new products and
services that are increasingly away from the
core business and that require rapid changes
in business processes and IT systems.
Technology Disruptions: Technology advances
i n comput i ng and net worki ng requi re
organizations to update their legacy systems.
With the explosion of ecommerce and mobility,
IT is the key enabler for new business models.
In a Forrester study, over a thousand
key decision makers from North America
and Europe were asked about their priorities
for from IT. Not surprisingly in the current
economic scenario, cost reduction (89%) and
improved business process execution (84%)
emerged as top business goals [2].
However, CXOs also recognize the fact
that to remain relevant and competitive in the
future, frms need to be more nimble to change
business processes (71%), improve collaboration
and information exchange using technology (69%).
Organizations are realizing enterprise
c ompl exi t y af f ec t s an or gani zat i on s
productivity thereby affecting profitability
and impairing competitive ability.
SMARTER ORGANIZATIONS OVERVIEW
The t hree charact eri st i cs of a smart er
organization is the ability to simplify IT
systems, learn and collaborate better and to be
able to adapt quickly [Fig. 1].
Simplify: By simplifcation of business processes
and IT systems, accurate and faster information
fow within the organization can be achieved.
Identifying key application portfolios,
creating shared services model by retiring
duplicate systems within the organization
translates to immediate business results in
effciency terms.
Data is a strategic resource which all
organizations possess but few are able to
leverage effectively. Analytics play a key role
in understanding vast reservoirs of data to
predict client behavior trends and optimize
supply costs.
Learn and Collaborate: Global organizations
today have a diverse workforce in terms of
culture, demographics and geographies.
In order to ensure efforts are not
duplicated and information exchange is speedy,
the traditional learning methods are no longer
effective for a mobile workforce.
Simplify
Underlying IT
systems to ensure
accurate & faster
information flow
Adapt
Be able to track and
optimize business
objectives on a real
time basis.
Take fact based decisions
faster in an uncertain
business environment
Learn &
Collaborate
Better knowledge
exchange
in a distributed
multi-cultural
workforce
Figure 1: Characteristics of Smarter Organization
Source: Infosys Research
17
Knowledge management processes and
tools gain greater credence to achieve better
information exchange and collaboration to
foster a culture of innovation.
Adapt: A smarter organization will be able
to define and track business objectives on
a real time basis. By providing customized
view of critical information to each business
user, proactive interventions and fact based
decision making can be enabled thus making
the organization nimble and smarter.
IMPACT OF CLOUD COMPUTING
In the past two years, there has been a
great spurt in interest in cloud computing.
Looking beyond the hype, CIOs need to ensure
investments in cloud based offerings deliver
IT simplification and significant reduction in
total cost of ownership. The key concerns that
remain to be completely resolved include data
security, scalability and potential lock-in with
a single vendor.
Current offerings can be classified under three
broad offerings as explained below.

Infrastructure as a Service (IaaS): In IaaS,
clients hire raw computing power such as
CPU, disk storage and network resources.
IaaS analysis must include not only the cost
of the computing resources but also the whole
operational and management environment
required.
Platform as a Service (PaaS): Paas involves
pay-per-use for an appl i cati on pl atform
which provides tools to develop and maintain
specific business applications. PaaS should be
considered only if significant benefits on cost,
flexibility, development effort and life cycle
management can be obtained as compared to
other options including in-house development.
Software as a Service (SaaS): SaaS involves
hiring a ready-to-use hosted product that
requires minimum or no customization to
meet business requirements. Selecting a SaaS
solution requires more attention as compared to
other cloud services as a best-of-breed solution
might fit the business need but may result in
creating new silos in the IT landscape.
IMPACT OF ENTERPRISE INFORMATION
MANAGEMENT
There is a clear shift from the traditional
reporting mindset to more analytical approach
to enterprise information. The second step
is assimilating structured and unstructured
information and delivering key insights to
stakeholders on a timely basis. There are
several areas that need to be addressed.
Master Data Management (MDM): To specify
set of processes and tools that defnes and manages
key data entities to ensure a single version of truth.
ERP Systems: Several organizations have
ERP/CRM packages that are rich sources of
critical information. However, inter-operability
i ssues l ead to data non-avai l abi l i ty and
inconsistencies.
Data Warehousing/BI: Legacy systems that are
focused on transaction reporting and which
require a strategic re-focus to deliver predictive
capabilities to aid decision making.
Enterprise Content Management: Provides
processes and tools to store and organize
unstructured data, for e.g., documents, wikis,
emails and information portals.
18
Enterprise Search: It is the process of making
available information from various sources
searchable for given audience. This would
require strict access management to ensure data
confdentiality.
Event Capture: Unexpected or atypical actions
by signifcant volume of customers or vendors
should be captured and escalated quickly to
senior stakeholders. This is more strategic for
B2C frms, for e.g., banking, retail and services
to take proactive steps to analyze and respond
faster.
ROLE OF SYSTEM INTEGRATORS
SIs today possess the business and technology
skills to build a comprehensive enterprise
architecture that will help align IT investments
with a long term business strategy and help
reduce organizational complexity by making
applications interoperable. Also they are
capable of integrating emerging disruptive
technologies while keeping total cost of
ownership under check.
There are four sub-architectures of enterprise
architecture, viz.,
Business: Business sub-architecture defines
the business strategy, model, processes and
the foundation for the other sub-architectures.
Information: Information defines the data
needs of the organization. It identifies and
standardizes information, assigns ownership
and accountability. The data exchange formats
are also defned in this sub-architecture.
Application: Application sub-architecture
defines the application design patterns and
enabling technologies. It also provides an
overview on how the applications should
be bundled to support a business process.
Business Architecture
describes the business
strategy, models,
processes, services and
organisation. Provides the
other enterprise architectue
dimension base their
decisions
Technical
Architecture
Application
Architecture
Business
Architecture
Information
Architecture
Information Architecture identifies,
documents and manages
the information needs of the
enterprise, assigns
ownership and accountability
for this information and
describes how data is
stored by and exchaged
between stakeholders
Application Architecture
defines the specification
of technology enabled
solutions in support of the
business architecture.
Provides a view on how
services should be bundled
to support a business
process
Technical Architecture
defines the strategies and
standards for technologies
and methods used to develop,
execute and operate the
application architecture.
It provides frameworks, technical
patterns and services that support
application requirements
Figure 2: Components of Enterprise Architecture Source: Infosys Research
19
This also specifes the middleware and other
software applications.
Technical: Technical sub-architecture specifies
the strategies and standards for underlying
netwwork and hardware that supports the
application and information sub-layers. This
sub-architecture lays down guidelines for
data.
Business Advisory Skills: Beyond technology
capabilities, strong understanding of business
processes and people management is very
critical. Change management is often an
underestimated factor in large transformational
initiatives that often leads to failure.
CRITICAL STEPS IN EXECUTION
In order to succeed in the journey to become a
smarter organization, following steps must be
understood and executed.
Right Mix in Leadership: Recognizing the
transformational nature of this journey, the
leadership must include mix of business
and technology leaders to define the road-
map.
Mission and Goals: A clearly articulated
mission and goals in both business and IT
terms should be published and communicated
to all stakeholders. Periodic update meetings
to review progress and update the roadmap
are important.
Business-led Metrics: Align the IT metrics and
milestones with clear business goals. For e.g.,
10% increase in new customer acquisition, or
15% cap on custom applications. These new
metrics will ensure IT architecture will be
tightly aligned to business outcomes.
Continued Executive Sponsorship: It is essential
to understand that transformation is an
ongoing exercise that requires periodic reviews,
assessment and re-alignment of goals and
resources. Senior management must commit
themselves to engage with IT leadership and
vendors on an ongoing basis.
WORKING IN MULTI VENDOR SCENARIO
Increasingly clients are choosing to work
wi t h f ewer st rat egi c vendors on l arge
transformational initiatives. While this results
in reduced total cost of ownership (TCO) and
reduced program management efforts, it places
greater responsibilities on system integrators to
demonstrate depth in capabilities and resource
scalability. Vendor selection and appropriate
governance mechanism needs to be put into place.
It is essential for key IT initiatives to be linked
clearly with business objectives. By creating
responsibility matrix for all stakeholders such
as RACI (Responsible, Accountable, Consulted,
Informed), vendors and client stakeholders
can work closely on shared understanding of
business goals and IT priorities.
CONCLUSION
Tradi t i onal l y IT was vi ewed a support
function and was dealt with tactically. Years of
inconsistent IT strategy and ad-hoc technology
adoptions leaves an organizations IT landscape
with several silos and ineffciencies. This has
led to adverse impact in business process
efficiencies and thus the ability to serve
customers better. The unwieldy IT systems erode
an organizations ability to compete effectively
in an uncertain business environment.
The new disruptive technology waves
like cloud computing and mobility can lead
to new technology silos and increase system
20
integration and maintenance costs in the future.
Therefore, for an organization to become
smarter it is essential to address enterprise
complexity, by building fexible, cost effective
IT systems, creating a learning environment
where employees, customers and vendors can
communicate and collaborate in real time. This
will help organizations adapt faster to key
organizational challenges.
In order to achieve these goals, smarter
organizations are increasingly investing in
enterprise architectures that have defined
bus i nes s obj ect i ves and s t r at egi es t o
streamline business processes and deliver on
organizational information needs. SIs today
have business and technology expertise to
help organizations simplify their IT landscape
and leverage technology to create a learning
environment.
REFERENCES
1. Gomolski, B. and Potter K. (2009),
Integrated IT Spending Perspectives,
Gar t ner Res ear c h. Avai l abl e at
h t t p : / / www. g a r t n e r . c o m/ i t /
content/1303700/1303722/march_17_
gar t ner _i nt egr at ed_i t _spendi ng_
perspectives_kpotter.pdf.
2. Kisker, H., et al. (2010), The State Of
Enterprise Software And Emerging
Trends: 2010 Forrester Research. Available
at http: //www. forrester. com/rb/
Research/state_of_enterprise_software_
and_emerging_trends/q/id/55802/t/2.
3. J anaki Akel l a, J . et al . ( 2009) , I T
Ar chi t ect ur e: Cut t i ng cost s and
Compl exi t y, Mcki nsey Quart erl y.
Available at http://www.mckinsey.
de/downloads/publikation/mck_on_
bt/2009/mck_on_bt_16_it_architecture.
pdf.
4. Heffner R., Cullen A., An M. (2010).
Top Pri ori ty For Cl oud Pl anni ng:
Integrate Cloud Computing Into Existing
Architecture Strategies. Dont Fall For
Hype Build A Strong Foundation
For Your Appr oac h To Cl oud.
Available at http://www.forrester.
com/rb/Research/top_priority_for_
cloud_planning_integrate_cloud/q/
id/56548/t/2.
5. Sohel, A., Obitz, T., Modi, R., Sarkar,
S.(2005). Enterprise Architecture: A
Governance Framework. Available at
http://www.infosys.com/consulting/
architecture-services/white-papers/
Documents/EA-governance-1.pdf.
6. White, M. and Briggs, B. (2010), Depth
Perception: A dozen technology trends
shaping business and IT. Available
a t h t t p: //www. de l o i t t e . c o m/
us/2010technologytrends.
7. Goel, M. (2009), Are you plagued with
unnecessary costs in your IT projects?
SETLabs Briefngs Journal, Vol 7 ,No
4 ,pp 4-10. Available at http://www.
infosys.com/infosys-labs/publications/
Documents/OLAR-project-costs.pdf
21
VOL 10 NO 2
2012
Effectiveness of EA Frameworks
in Achieving BITA:
An Analysis Based on SAMM
By Ramkumar Dargha
Infosys Labs Briefings
B
usi ness IT Al i gnment (BITA) i s one
of the mai n goal s of any enterpri se
architecture (EA) initiatives. EA frameworks
like ZACHMAN [1], TOGAF [2], and FEA [3]
help organizations in their EA journey and
are expected to help achieve BITA goals. But
questions that constantly get probed like (a)
how far the EA frameworks help in achieving
BITA; (b) what kind of support they provide and
what they do not; (c) whether they cover aspects
of aligning business to IT or IT to business, help
in assessing the utility and relevance of such
frameworks.
Among the many BITA effectiveness
measurement models available in literature,
the strategic alignment maturity assessment
model (SAMM) by Luftman et al [4] is the most
comprehensive model that is typically used to
assess the BITA alignment in an enterprise.
The 40+ factors provided in this model can
also be used as an assessment model to assess
the capability and support provide by other
EA frameworks for achieving BITA. This
paper attempts at applying the SAMM model
to the popular EA frameworks like TOGAF,
ZACHMAN, etc., to analyze the effectiveness
of the support provided by these frameworks
in achieving BITA.
BUSINESS AND IT ALIGNMENT
BITA is the degree to which the enterprise
business components composing of business
models, business strategy, processes and
business organization are aligned with IT
infrastructure, IT applications, IT architecture
and IT organization or vice versa. Traditionally,
BITA is the degree to which the enterprise
business components composing of business
models, business strategy, processes and
business organization are aligned with IT
22
BITA meant referred to IT aligning with business.
However, today enterprises have realized that
businesses too need to be aligned to IT. ITs
alignment with business or business alignment
with IT are not mutually exclusive. They co-
exist with alignment being the core point of
focus. SAMM allows one to analyze both the
perspectives. SAMM provides a set of criteria
that can be used to assess EA frameworks ability
and support to achieve BITA alignment. The
model captures different factors that determine
BITA effectiveness. SAMM organizes these
factors into six categories, viz., communication,
value measurement, governance, partnership,
scope and architecture, and skills maturity.
ANALYSIS OF EA FRAMEWORKS
USING SAMM MODEL
While there are multiple EA frameworks in
literature, it is important to restrict the scope
of this study to the most popular few to assess
if they are able to satisfy the different criteria
set by SAMM, thereby addressing the BITA
needs of an enterprise. For the purpose of this
study only a select few popular frameworks are
considered and they are analyzed against the
six criteria laid out of SAMM.
Communication Maturity
TOGAFs Architecture Development Method
(ADM) i s a popul ar method to devel op
enterprise architecture. In addition to ADM,
TOGAF provides clear definitions, content
frameworks, templates, matrices, artifacts,
guidelines, techniques and reference models
for different components of architectures.
TOGAFs capability framework provides
guidelines on processes, skills, roles and
responsibilities required for establishing and
operating architecture functional group within
an enterprise.
So how do all these help in meeting the
required communication criteria prescribed by
SAMM model for BITA?
Artifacts and repository created as
part of TOGAFs architecture vision phase
helps communicate the vision, mission, goals,
drivers, value propositions, constraints, key
performance indicators (KPIs), risks, mitigation
activities and high level business, information,
application and technology architecture of EA
initiative. Stockholder expectations capture
stakeholders, their concerns and requirements
that is one of the first steps in creating the
communication plan.
The set of standard architectural
content and deliverables like architectural
bui l di ng bl ocks, archi t ect ure cont ract ,
principles, requirements, roadmap, models,
change management plans, implementation
plan, compliance plan and other standard
architecture contents, as defned in TOGAF and
more elaborately in ZACHMAN framework,
provide a strong means of communication to
all stakeholders.
ZACHMAN frameworks specific
emphasis on architectural artifacts for different
perspectives/roles (like planner, owner,
designer, builder and subcontractor) and for
different aspects (what, how, where, who and
when) enables enterprises to create detailed
artifacts that links the needs of different roles
and lifecycle stages, and thus plays a key role in
helping IT understand business and vice versa.
The goal of FEA f ramework s
reference models is to facilitate communication,
cooperat i on, and col l aborat i on across
organizational boundaries.
Liaison effectiveness that is an important
factor in SAMMs communication criterion can
be incorporated as part of the EA architecture
governance board or architecture capability
23
group. However, effectiveness measurement
of such a liaison activity needs to be added
to the architecture performance measurement
framework.
The EA frameworks thus do a good job in
helping to achieve four of the six communication
needs as outlined by SAMM viz., understanding
of business by IT, understanding of IT by
business, inter/intra organization learning,
liaison effectiveness.
However, some gaps still exist in the three
frameworks, with respect to the remaining two
communications criteria. . For example, there
are no specifc guidelines on communication
protocol fexibility or protocol rigidity in the EA
frameworks. Also they do not elaborate much
on the communication plan aspects.
In addition, the frameworks do not
provide any specifc guidelines for knowledge
sharing. Knowledge sharing is an important
element in the communication criteria as
laid out in SAMM. A typical knowledge
management initiative should take into account
the following four aspects:
Technology: These are the enabling
technologies that help in knowledge
gathering, creation, storing, distribution
and enhancement.
People: Issues related to motivating
employees and other participants in
knowledge creation and sharing.
Processes: Issues related to knowledge
creation process, knowledge protection,
IP and legal aspects.
Culture: Issues related to cultivating
the overall organizational culture of
openness.
Value Measurements Maturity
It refers to the measurement of value that IT
as a whole brings to business. Measuring such
a value effectively and making improvements
based on such value measurements is one
of the main criteria to measure BITA as
per SAMM. So the question is that of how
EA frameworks through their guidelines,
processes and recommendations help in IT
value measurement.
TOGAF has a section on performance
measurement s wi t hi n t he archi t ect ure
governance module where EA performance
evaluation and business value evaluation of
such architecture is covered briefy. ZACHMAN
framework also mentions about the importance
of capturing business value achieved in the
cells corresponding why aspect and planner/
owner perspectives. But the EA frameworks
do not provide comprehensive guidelines
and recommendations on how to measure
the business value of IT as a whole. These EA
frameworks need to be supplemented with other
standard IT measurement frameworks like -
Business Vaue Index (BVI): Used by
enterprises that are new to IT value
measurements as initial steps to IT value
measurement [5].
Total Economic Impact (TEI): From
the Forrester stable, this is a stronger
quantitative framework compared to
BVI. A big advantage of this method
is that it measures intangibles such as
fexibility of IT as well [6].
Val IT: An IT Governance Institute
(ITGI) framework this is used mainly for
evaluating new investments rather than
current ongoing IT projects.
24
Appl i ed Informat i on Economi cs
(AEI) : Very quantitative in nature it
is extremely complex in its nature but
presents the most accurate net present
value (NPV) analysis [7].
SAMM recommends measuring business
value through a balanced metrics approach
which essentially is the combination
of business and IT metrics. The listed
standard IT measurement frameworks
when applied with EA frameworks help
achieve the value measurements criterion
as prescribed by SAMM.
Governance Maturity
There are multiple types of governance that
are of importance to an enterprise, viz. ,
corporate governance, IT governance and
architecture governance. SAMM deals with IT
and architecture governance aspects. The EA
frameworks like TOGAF and FEA deal mainly
with architecture governance. IT governance is
a much broader than architecture governance.
Forrester defines IT Governance as the
process by which decisions are made around
IT investments. How decisions are made? Who
makes the decisions? Who is held accountable,
how the results of decisions are measured and
monitored are all parts of IT governance? [8].
TOGAF has a comprehensive architecture
governance framework consisting of context,
process, content, repository, monitoring,
reporting, structure, key success factors and
compliance guidelines. But the EA frameworks
do not have similar coverage for IT governance
and IT service level governance.
Enterprises have to use additional
governance frameworks like COBIT, ITIL and
COSO to meet the overall IT governance needs
[9, 10, 11].
C O B I T p r o v i d e s t h e mo s t
comprehensive governance framework with a
strong focus on process of governance, strong
auditing and controls perspectives. While
ITIL provides a framework for IT service
management, COBIT takes the perspective of
audit and control.
Other governance frameworks like
Forrester framework organize governance
around three main categories, viz., governance
st ruct ures, governance processes, and
governance communication. While the last
two aspects are addressed in COBIT and ITIL
to a certain degree, the Forrester framework
provides additional emphasis on governance
structures and reporting structures.
By using above frameworks in unison
with EA frameworks the enterprises would be
able to satisfy governance needs as set forth
by SAMM.
Partnership Maturity
The EA frameworks do not deal with business-
IT partnership model or guidelines explicitly.
But the EA frameworks implicitly capture
some of the important aspects required for
an effective business-IT partnership. Before
analyzing EA frameworks support for such a
relationship, let us look at some of the typical
key ingredients of a good partnership model.
A partnership model is typically comprised
of some key ingredients like (a) drivers that
are a compelling reason to partner and that
set expectations on outcomes; (b) components
that joint activities and processes that build
and sustain the partnership; (c) facilitators
that enhance partnership growth; (d) outcomes
that refer to the extent performance meets
expectations; and (e) feedback that essentially
refers to the inputs given by partners for
continuous improvement.
25
Let us now analyze EA frameworks
support i n i mpl ement i ng a successf ul
partnership model.
Drivers: This aspect is actually an enterprise
level responsibility and is out of scope for a
typical EA framework. These drivers would
be set before venturing into an EA program.
Components: The EA frameworks propose
and recommend many activities and processes
that not only help in creating a good EA, but
indirectly help in building and sustaining
business-IT partnership. In TOGAFs ADM
model, almost all activities have both business
and IT aspects. For example, the stakeholder
lists for each activity and deliverable include IT
and business representatives, the architecture
governance board and architecture center of
excellence (COE) will have representatives
from both IT and business. In addition, the
archi tecture i mpl ementati on starts wi th
business architecture with due importance given
to linking business architecture to subsequent
IT architectural components like information,
application and technology architectures.
Finally, business viewpoints are included in
each of the IT architecture artifacts as proposed
by TOGAF and ZACHMAN drives business-IT
collaboration. All these contribute to the core
components on which a business-IT partnership
can prosper and function effectively.
Facilitators: The EA frameworks propose and
recommend setting up of architecture governance
board and architecture capability function or
COE. They also recommend representation of
both business and IT groups in these boards and
COEs. These boards/COEs and in addition to
focused/dedicated liaison teams can perform
the role of facilitators in building trust and
partnership between business and IT.
Outcomes: The most important outcome
of effective business-IT partnership would
be to achieve the goals of IT as required by
business. Though EA frameworks themselves
do not cover the metrics and performance
measurement in great detail, there are other
additional frameworks that can supplement to
achieve the outcomes.
Feedback: Facilitators and outcomes play an
important role in the feedback mechanism.
In addition, the stakeholder communication
section in the communication plan needs
to incorporate required formal feedback
mechanisms for IT-business partnership for the
purpose of continuous improvements.
As can be s een f r om t he above
analysis, though the EA frameworks do not
explicitly address the business-IT partnership
needs, there are inherent mechanisms as
recommended by t hese EA f rameworks
that can be utilized for such an effective
partnership model.
So how do al l t hese mat ch up t o
SAMMs maturity criteria? The combination
of components, facilitators, outcomes and
feedback together can help improve business
perception of IT value and can enhance mutual
trust resulting in shared goals, risks and
rewards, all of which are important criteria in
SAAM for achieving an effective business-IT
partnership model.
Scope and Architecture Maturity
SAMMs scope and architecture component
measures ITs ability to create a flexible
infrastructure, its ability to evaluate and apply
emerging technologies, its ability to enable
or drive business process changes, and its
delivery of valuable customized solutions to
internal business units and external customers
26
or partners [12]. This component answers
some of the questions like: To what extent has
IT evolved to become more than just business
support? How has IT helped the business
to grow, compete and proft? The criterion
covers two important items: (i) ITs ability to
create fexible IT systems to satisfy the ever
changing business needs in an agile and effcient
manner; and (ii) ITs ability to perform the role
of a business enabler or driver rather than
just as business support or business process
automation function.
The first aspect above is addressed
to a great extent in EA frameworks. For
instance, both TOGAF and FEA EA frameworks
l ay si gni fi cant emphasi s on technol ogy,
information, security and implementation
standards. In all EA deliverables prescribed
by TOGAF, compliance to standards is given a
very important role. TOGAF includes separate
sections on technologies and frameworks that
enable implementation of a standard and fexible
architecture. Standards information base (SIB)
is one of the main components of TOGAFs
architecture repository. The SIB captures the
standards with which new architectures must
comply, which may include industry standards,
selected products and services from suppliers,
or shared services already deployed within the
organization. TOGAF also prescribes using
other related standard frameworks like ITIL,
CMMI, COBIT, PRINCE2, PMBOK as applicable
in different phases of EA implementation
cycle[13,14].
FEA t oo recommends appl i cat i on
development, technology implementation,
project management, governance, vendor
management, production operation, architecture
governance, confguration management and
problem resolution in every phase of EA
lifecycle implementation.
Standards and flexible architectures
prescribed by EA frameworks provide a
mechanism to create flexible and standards
based IT infrastructures that are more amenable
for implementing an agile and effcient system. .
Standard and fexible architectures satisfy
SAMMs criteria like standard articulation,
archi t ect ural i nt egrat i on, archi t ect ural
transparency and fexibility.
Skills Maturity
EA frameworks provide a comprehensive
coverage on the skills defnition and depth of
knowledge required by different roles in an EA
team. TOGAF covers skills from different skill
dimensions viz., generic skills, business skills,
EA skills, program/project management skills,
IT general knowledge skills, technical IT skills
and legal environment skills) and also covers
different roles viz., architect board member,
architect sponsor, EA manager, technology/
data/application architects, managers and IT
designers in each of the dimensions.
B y f o l l o wi n g t h e g u i d e l i n e s a n d
recommendations prescribed by TOGAF
as above, an enterprise would be able to
satisfy most of the skills criteria set forth by
SAMM. But few other criteria like innovation,
entrepreneurship, locus of power, trusting
envi ronment and career crossover have
more to do with generic organization skills/
capabilities, for which enterprises need to
leverage standard practices available in
organizational management literature.
CONCLUSIONS
A thorough evaluation of EA frameworks
against the canvas of SAMM suggests that
they are riddled with the lacunae of being
reactive frameworks. Hence, it would do
well for an enterprise architect to adopt an
27
extended framework that consists of the
current EA frameworks along with some key
supplementary frameworks and that meet the
criteria set forth by SAMM to achieve fawless
BITA [Fig 1].
Using this extended framework, an enterprise
can proactively take measures and employ
practices to satisfy BITA needs rather than frst
discovering gaps and then taking measures
while implementing an EA initiative.
Compared to TOGAF, ZACHMAN and
the like, it is more comprehensive and enhances
the repertoire of tools that an EA consultant can
have before embarking on an EA engagement.
It defnes a pre-defned set of tools/frameworks
that an EA consultant can pick and choose
during an EA engagement. This increases the
predictability of success of an EA engagement
and reduces time to get started on such an
engagement.
It must however be understood that the
extended framework does not re-invent the
wheel but uses the existing frameworks like
TOGAF, ZACHMAN and builds on top of these
frameworks to address the gaps in order to meet
the BITA needs.
REFERENCES
1. The Zachman Framework. Available
at http: //en. wi ki pedi a. org/wi ki /
Zachman_Framework
2. TOGAF Framework. Available at http://
www.opengroup.org/togaf/
3. FEA Enterprise Architecture. Available at
http://en.wikipedia.org/wiki/Federal_
Enterprise_Architecture
4. Luftman J., (2000) et al, Assessing Business
IT alignment Maturity, Communications
of the Association for Information Systems
AIS, Vol. 4, Issue 14. Available at http://
aisel.aisnet.org/cais/vol4/iss1/14
Business IT
Alignment
EA Frameworks
EA Frameworks
Skills
Organization Learning
Model
Innovation Management
Model
Scope and Architecture
IT driven BPR frameworks
Extended BPR frameworks
Industry Case Studies
Communication
Knowledge
Management Tools
Specialist Liaison Groups
Flexible Protocol Structures
Value Measurements
Business Value Index
Val IT Model
TEI Model
AIE Model
IT Balanced Scorecard
Partnerships
Trust based Strategic
Partnership Models
Cross Participation
Team Effectiveness
Models
Governance
COBIT
ITIL
COSO
ISO 17799
Figure 1: Extended BITA Framework Source: Infosys Research
28
5. Business Value Index. Available at
http://download.intel.com/it/pdf/
ITBusinessValue.pdf
6. TEI- Total Economic Impact. Available at
http://www.forrester.com/TEI
7. AEI- Applied Information Economics
Available at http://en.wikipedia.org/
wiki/Applied_information_economics
8. Symons, C., (April, 2006), Topic Overview
IT Governance. Available at http://
www. forrester. com/rb/Research/
topi c_overvi ew_i t_governance/q/
id/39177/t/2
9. COBIT: IT Governance Framework.
Available at http://www.isaca.org/
Knowledge-Center/COBIT/Pages/
Overview.aspx
10. Information Technology Infrastructure
L i b r a r y ( I T I L ) . Av a i l a b l e a t
http://www.itil-offcialsite.com/
11. COSO: Commi t t ee of Sponsori ng
Or gani zat i ons of t he Tr eadway
C o mmi s s i o n . A v a i l a b l e a t
http://www.coso.org/
12. S AMM: S o f t wa r e As s u r a n c e
Ma t ur i t y Mo de l . Ava i l a bl e a t
http://www.opensamm.org
13. PRINCE2: PRoj ects IN Control l ed
Envi r o nme nt s 2 . Ava i l a bl e a t
www.prince-offcialsite.com/
14. PMBOK Guide: Project Management
Body of Knowledge. Available at www.
pmi.org/PMBOK-Guide-and-Standards.
aspx.
29
VOL 10 NO 2
2012
Proactive performance engineering holds the
key to improving performance and scalability
requirements in a modernization initiative
Proactive Performance
Engineering for
E-commerce Modernization
By Saurabh Kumar Mishra, Balaji Subbaraman and Rohit Christopher
Infosys Labs Briefings
T
oday, enterpri ses are embarki ng on
major transformational IT programs to
improve customer experience, to cater to
increasing volumes, to ward off competition
by innovating on products and services, in
addition to reacting swiftly to newer and newer
business requirements. They are increasingly
adopting newer technologies and architectural
trends to modernize their business platforms.
Modernization of existing platforms has in
a way become the need of the hour and no
sane enterprise can go about engaging in a
modernization initiative minus a clearly defned
methodology.
In this paper, we try to articulate a simple
approach which if followed would ensure the
success of the modernization program. For
the sake of focus we dwell on ecommerce
modernization. The approach though applies
to any other modernization initiative as well.
NEED FOR E-COMMERCE MODERNIZATION
E-commerce sites differ vastly in comparison
to traditional brick and mortar shopping sites
in that they offer a way too different shopping
experience to the customers. For marketers, the
quintessential question remains that of clocking
a higher basket-to-order conversion. In a bid
to improve on this conversion, the ecommerce
application must be so developed that it is
easy to navigate, should allow for product
customization, must have a very simplified
checkout process and also be robust in analyzing
the site visit patterns of the visitor to be able to
send her personalized promotional offers.
It is common knowledge that in the
recent times online buying constitutes a bulk
of a customers spends. How do e-commerce
sites then leverage this information to scale up
their online sales? Personalization, navigability
and order fulfillment are three pillars on
30
which many e-commerce sites have built their
businesses. Hence there is no gainsaying the
fact that these three pillars constitute the
most important elements in the success of
e-commerce.
Personalization E-commerce si tes use
specialized software that help create unique
ser vi ce bout i ques t hat t ar get speci f i c
customers. There are exampl es abound
that go on to demonstrate the importance
of personalization. For example, American
Airlines employs filters that classify types of
travelers and customizes its offerings based
on such classification.
Easy to use Application An easy to use,
interactive website is the key to entertaining
a customers online shopping experience. This
should be backed with good support via phone,
live chat and or mobile shopping.
Fulfllment E-commerce has increased the
focus on customer satisfaction by offering on-time
delivery fulfllment and various options for pickup
like store pickup with no extra delivery charges.
These features are likely to see some
innovation in the years to come and it is
important for e-commerce enterprises to embrace
modernization to keep brace with anticipated
growth in customer requirements and more
importantly to keep pace with infrastructure,
security and scalability requirements. Proactive
performance engineering can help counter the
huge challenges that any such modernization
initiative has to offer.
PROACTIVE PERFORMANCE ENGINEERING
Performance engineering is a specialized
technical stream of work that requires a
practitioner to have in-depth knowledge
on various technology areas and expertise
t o archi t ect and desi gn hi ghl y scal abl e
applications that should meet performance
objectives along with the ability to identify,
analyze and provide solutions. It requires
subject matter experts in areas like architecture,
application, infrastructure and products across
different technologies.
Ma ny pe r f o r ma nc e i s s ue s a r e
identified only during the non-functional
testing phase of any software development
life cycle (SDLC). Maj ority performance
improvement solutions involve either making
a considerable amount of code changes or
design changes that are characterized by time,
cost and effort drags. For many applications,
go/no-go live decision depends purely on
their performance and unfortunately such
perf ormance i s real i zed j ust bef ore t he
applications are launched.
I n mos t s cenar i os , appl i cat i ons
are desi gned by consi deri ng functi onal
requirements rather than non functional
requirements such as performance & scalability.
To address this concern it is important to
adopt proactive performance engineering in
transformation programs that help in focusing
on performance objectives at every stage of the
development.
To continuously ensure application
performance and its scalability, it is important
for proactive performance engineering to
engage in activities like
(a) Identifying business critical transactions
using analytical processes;
(b) Capturi ng busi ness vol umes and
expect ed response t i mes duri ng
requirements analysis phase;
31
(c) Reviewing architecture and application
designs from performance and scalability
viewpoint;
(d) Conducting proofs of concept to ensure
that the proposed solution or an or an
architecture component or design can
sustain the expected volumes;
(e) Ensuring optimized code development
that is achieved by automated and
manual code reviews and profling of
critical transactions;
(f) Developing performance models using
industry standards such as queuing
net work model ( QNM) and ot her
relevant techniques.
METHODOLOGY
The methodology used for implementing the
proactive performance engineering approach
can be defned as that of evaluating performance
objectives at each and every stage of the
application development lifecycle.
T y p i c a l l y , l a r g e e c o mme r c e
modernization programs involve various
functi onal modul es such as ecommerce
website, order management system, product
catalog, search engine, merchandising, content
management, enterprise service bus (ESB),
customer support, personalization and other
modules. Currently, no single product provides
a comprehensive solution that meets all of the
above features for very high concurrent user
sites and large product volumes. This leads to
building a solution with a mix of products and
Figure 1: Stages in Proactive Performance Engineering
Methodology
Source: Infosys Research
Design Phase Build & System Testing Phase E2E Testing Phase Post Production
Low level design recommendations reports
Code review reports
Performance test strategy
Measurement strategy
Measurement logs & Analysis reports
Deliverables across Phases
Deliverables
Identification of Critical Transactions
Workload modeling & Transaction profiling
Forecasting Loads
Kite and Key note data analysis
Use case - Interface dependency analysis
Baseline performance requirements
Endorsement from product vendors on
NFRs and rationalization of NFRs
Design and Architecture Reviews
Identify and Execute POCS
Create software execution models
Low level design review
Manual code review of critical components
Identifying PMD rules to enable code
review automation
Performance test strategy creation
Code profiling
Component & Transaction performance
testing
Update Software execution Models
Define performance measurement strategy
Review environment
Provision tools
Metrics Analysis
Identify optimization areas
Optimize environment & code
Revisit capacity plans
Monitoring the key Performance Indices
Quick resolution of performance issues
Impact Analysis of CRs
P
e
r
f
o
r
m
a
n
c
e

R
e
q
u
i
r
e
m
e
n
t
s
Performance Engineering Activites
Workload model, Requirements & Forecasts
Proof of concepts
Design reviews and recommendations
Software execution model
32
technologies that need to be integrated. This
imposes tremendous complexities in measuring
and achieving performance objectives as each
product/technology has its own benchmark
data that can impact the overall integrated
solution performance.
It is therefore important to have a
dedicated team to look after the performance
objectives across the solution. The primary
focus of the team should be that of defning
perf ormance obj ect i ves, governi ng t he
implementation process and defning gating
criteria at every stage of implementation that
passes through the performance validation.
Some key steps that are crucial to
rolling out a sturdy e-commerce modernization
program are discussed below.
Identifcation of Critical Transactions
Traditionally, critical transactions of existing
applications were identified through log
analysis and usage statistics. However, with
subjectivity taking center stage, with each
business team providing its own views without
factoring in the technical complexities, the old
system of transaction identifcation has become
redundant. Today, both business as well as
technical views are to be reckoned in identifying
critical transactions. This can be done with the
architects articulating the technical complexities
in the application and the business stakeholders
rat i ng t he t ransact i ons based on some
pre-defned business criteria.
Workload Modeling
The next step is that of base-lining workload
requirements of the proposed system to
validate application scalability requirements
for future. This activity of base-lining workload
requirements is popularly known as workload
modeling which identifies different forces
acting on the system such as session length,
concurrency, average and peak loads, arrival
patterns, and current response.
Wor kl oad char act er i s t i cs ar e t o
be derived through log files generated at
application layer, web server layer and database
layer for the existing systems. The critical
transaction identified will be the input for
workload modeling which will identify the
online transaction processings (OLTP) page
response time and the load it can sustain in
future.
Designing and Reviewing Architecture
Performance
Application architects provide design and
architecture solutions using best products,
technologies and architecture. However,
some of these decisions may not comply with
performance and scalability criteria. The
performance architects review the design
and architecture using methodologies like
Archi tecture Tradeoff Anal ysi s Method
(ATAM).
Each t ransact i on i s uni que i n i t s
operation and involves multiple http requests,
database calls, jms calls, remote invocation, xa
transaction, etc for a successful operation. The
review qualitatively measures the architecture
and design to respond within acceptable
response times for each of these touch points.
It factors into the evaluation the underlying
data volumes, concurrent requests, physical
and logical designs, underlying technologies
for the different operations to identify issues,
if any.
If the architecture and/or design do not
meet the required performance requirements,
performance architects should discuss the need
to modify the architecture, design and business
process to ensure that expectations are met.
33
Automating Code Review
It is important to defne automated code review
rules that help in building good quality code that
can scale. Also important is setting up of a gating
criteria to pass the automated code review
before it goes to test or build/release process.
Identifying and Executing POCs
During design stage, there may be various
solutions to address a problem statement, but
it is very important to pick a solution that can
meet the performance requirements. This might
require building a prototype and validating
the scalability nature of the proposed solution.
Conducting a proof of concept with reasonable
amount load testing on the proposed solution
will help select the right solution that can meet
the service level agreement (SLA) requirement
and scalability aspects.
Performance Modeling
Performance modeling enables one to abstract
the performance-related behavior of the system.
There are various tools and techniques that
help build a performance model. Using a
performance model one can predict
(a) Response time, both for the overall
system and per component;
(b) Expect ed wor kl oad on di f f er ent
infrastructure components like CPU,
IO, network, etc;
(c) The capacity needed for each class
of servers, to a certai n degree of
accuracy.
Validating Response Time
Validating response time is necessitated to out
if the current set of products in the context of
architecture is able to give the response time
and throughput. Performance model can be
used to predict response time in the production
environment.
CONCLUSION
Proactive performance engineering can be
very instrumental in successfully rolling out
a modernized application. It can identify
and resolve several issues all through the
application development lifecycle. These
include performance issues in out-of-the-
box product features of several products,
capt uri ng i ncorrect archi t ect ure desi gn
decisions at design stage and capturing code
level performance issues early on failing
which there can be considerable delays in
application rollout and concomitant business
losses.
REFERENCES
1. What is Kite? Available at http://kite.
keynote.com/faq.php.
2. Challenges in B2C was captured from
ht t p: //www. ecos vi r t ual es . com/
tutoriales/new/B2C.doc.
3. Paliwal, S. (2009), Interview: Getting
To Know Raj inder Gandotra, Head
of Perf ormance Engi neeri ng and
Enhancement (PE2) Practice at Infosys.
Available athttp: //www. cmg. org/
measureit/issues/mit62/m_62_9.html.
4. Architecture Tradeoff Analysis Method
(ATAM) http: //en. wikipedia. org/
wiki/Architecture_Tradeoff_Analysis_
Method.
5. Analytical Hierarchy Process(AHP).
ht t p: //e n. wi ki pe di a. or g/wi ki /
Analytic_Hierarchy_Process.
6. A n o v e r v i e w o f Q u e u e i n g
Ne t wo r k Mo de l . Ava i l a b l e a t
34
ht t p: //www. cs. washi ngt on. edu/
home s /l a z ows ka /qs p/I ma ge s /
Chap_01.pdf.
7. Sassenburg, H., Vioncea, L. (2008),
Modernization of Legacy Systems:
When and How?. Available at http://
www. s o l i ds o ur c e i t . c o m/f i l e s /
Modernization_of_Legacy_Systems.
pdf.
8. Simon, L.and Johnny W.(1982), Queueing
Network Models of Packet Switching
Networks, Part 1: Open. Networks,
Performance Evaluation, Vol. 2, No. 1.
9. Simon, L.and Johnny W.(1982), Queueing
Network Models of Packet Switching
Networks, Part 2: Networks wi th
Population Size Constraints, Performance
Evaluation, Vol. 2, No. 3.
35
VOL 10 NO 2
2012
SOA and BPM can be effectively employed in
an organizations transformation exercise
How Systems Integration
Helps Achieve Progressive
Business Transformations?
By Anirban Ghosal and Narayanan Chathanur
Infosys Labs Briefings
B
usiness transformation is aimed to produce an
enterprise level performance improvement
through radical changes in the business and IT
systems. There are a multitude of challenges
and dimensions to such programs. A few typical
questions that illustrate these challenges are:
Is transformation technology driven or
business driven? How to marry them in
any case?
Is the implementation phase-wise and
if so is it region-wise or unit-wise or
system-wise?
How to absorb any change due to
unforeseen factors such as split, mergers
or acquisitions?
How to put together a technology platform
that is durable, resilient and tolerant?
How to measure and apply course
corrections if things go off track?
This paper attempts at presenting a view
on how this complex process of progressive
transformation can be achieved through simple
steps aligning the business vision and IT vision
with the help of service oriented architecture
(SOA) and business process management (BPM)
methodologies.
Bus i nes s t r ans f or mat i on i n our
view is an endeavor aimed at producing
ambi t i ous measurabl e st ep- changes i n
business performance. The goals of a business
transformation endeavor are primarily around
(a) repositioning the company in the minds of
customers, shareholders and even the industry;
(b) innovating to take advantage of new sources
of competitive advantage; (c) realizing speed
in product and service delivery; (d) improving
business agility and flexibility to address
36
changing competitive conditions; (e) setting
new standards in product and service quality;
and (f) redesigning cost structures to improve
profitability and free up funds for top-line
enhancement activities
Business transformation results into
enterprise level performance improvement,
impacts the enterprise through changes in
the core processes, organizational culture,
operating structure and business performance
accountabilities. Transformation is typically
executed over a period of two to three years and
executed through a progressive transformation
approach. During the transformation new and
old business process as well as IT systems
coexist to support the business needs.
IT DRIVEN TRANSFORMATION
A GOOD CHOICE?
Currently, large transformation programs
are executed with a combination of turnkey
projects by multiple IT service vendors. These
vendors very often use their own methodology
for solution delivery. The role of enterprise
architecture (EA) is mostly restricted to defning
the new system landscape and new business
landscape and occasional review of various
IT systems in functional architecture, data
architecture and technical architecture spaces.
Return on investment (ROI) is not clearly
justifed in such cases and generally there is
more expectation from EA.
Enterprise architects on the other hand
have the constraints of not infuencing the business
leaders who are basically the stakeholders for the
IT systems supporting their business units or
processes. Such problems are prevalent in most
of the organizations as the IT services were not
conceptualized as part of an EA view.
In our opinion instead of the architecture
being driven by business, IT should drive with
a well-defined EA and its subset systems
integration architecture (SIA) at the beginning
of the transformation. This is the key to success
for business transformation since enterprise
transformation engrosses all the elements of an
enterprise. It is imperative for organizations
to ensure EA has a vision of future state
that can support the business that business
leaders envision for and have a clear vision
for the intermediate stages to ensure a smooth
migration to the new state. At the same time
EA should not have any qualms in aligning
their vision with business vision and business
expansion roadmap. Any gap between the
understanding of EA vision and business
vision, needless to say, will create bottlenecks
for both.
An important transformation objective
is that of business process optimization.
More often than not, products and services
can be imitated, but an organizations main
differentiator is its business process. Often
business process is where real innovation can be
created and how a company can gain signifcant
competitive advantages. Business process
automation and optimization beefs up this cause
and also is critical for business transformation.
Bottom-line business transformation driven by
business is to increase business effectiveness
that basically impacts the IT systems to improve
the optimization and automation. Second,
business transformation driven by IT is also for
the same cause but if there is any difference in
the way other perceives the end vision of the
transformation, it will not be a success. To
fortify the thought vision needs to be aligned
from both these angles.
SOA A GOOD OPTION
Since the fnal results of transformation are
visible to end users only at the end of journey,
37
the most obvious way organizations plan is to
engage in phase-wise implementation which is
one of the best ways to go through this journey
but the milestones should be time-bound and
not bound on delivery size. Let us illustrate this
contention through an example.
An organization having multinational
operations decides to create a new application
for their supply chain operations that require
integration with legacy systems, new services to
be created and change in the existing business
processes to optimize operations. Since the
scale of implementation was huge and was
planned for years, the management decided
to implement the first pilot for Asia-Pacific
region that had just about 10% of its user base.
Five other such phases were conceived to roll
out the initiative across other geographies like
A Rest of Asia, Europe, Africa, North America
and South America.
The implicit assumption in rolling out a
pilot is that setting up the frst platform is cost
and effort intensive and once the platform is
built it would be easier to reuse it multiple times.
While the plan was to deliver the platform by
the end of frst year, implementation took more
than two years due to bottlenecks like migration,
region-wise fltered data requirements, delays
due to involvement of multiple stakeholders,
time management, feedback management that
involved getting inputs on business process
design and usability, and above all retiring
legacy systems. Transformation efforts are
dotted with similar features in many such
turnkey project implementations.
Time-bound deliveries reduce such
risks and improve confdence. So instead of
creating all the functionalities for Asia-Pacifc
region, the plan should be to create independent
functions, services and integration fows that
could be delivered in much shorter periods,
say for example three months. This would also
ensure that every quarter there is a delivery
for business and issues if any are assimilated
more quickly in a production environment
and corrective action taken in the subsequent
quarters. Any course correction is also time
bound and helps the overall cause which
compensates for any rise in overhead costs.
Having an SOA platform thus helps in going
through this journey as by nature SOA is open
standard, progressive and reusable.
Because of progressive nature of the
transformation program that involves phase
wise migration of business processes and the
systems, SOA based approach must be adopted
in SIA for the organization [1]. SOA being a
pattern that helps in progressive adoption of
any technology into SOA bandwagon defnitely
helps in any kind of transformation whether
it is migration from legacy to modernization
or change in the way business process itself is
implemented. SOA also is not specifc to any
technology or product suite and adhering to
SOA standards by all the implementations
for the transformation program automatically
creates a robust network of services. EAs
vision should be to create the right kind of
SOA framework and compositions that suits
the business objective.
BPM OPTIMIZATION AND AUTOMATION
PLATFORM
Business processes needs to be reengineered
during and after business-IT transformation
to meet the ever changing business needs.
Transformational business processes need to
provide agility and fexibility to derive optimal
outputs. This is a minimum required of any well
designed business process. Business processes not
only need to be automated to bring agility but also
are to be optimized on a continuous basis.
38
From the business process point of view, BPM
capability brings the discipline into the business
process modeling and helps the enterprise
manage these processes more effectively.
Business processes are modeled, simulated,
and optimized through the BPM platform.
BPM platform manages the full life cycle of the
enterprise business process. It also increases
visibility into the business processes and
enables companies transform those processes
to achieve the desired results [2].
Business processes must be automated
and optimized through a centralized platform
in order to provide an effcient, effective and
reusable set of business processes across
the business units in an enterprise. The
consolidation of having these processes in a
centralized platform also helps in rationalizing
the cost towards implementing the tools and
technology underlying BPM in an EA.
IT landscape consolidation is one of the
key strategies in of business transformation
programs. Business processes are implemented
using the capability of the new generation as
well as legacy enterprise systems. The key
challenge with EA is to defne the role of these
enterprise systems and the BPM platform in
order to deliver the outcome of the business
processes effectively.
Illustrated in Figure 1 is a simplifed
Telecom business process. At the highest level
in this process there are Operations Support
Systems/ Business Support Systems (OSS/BSS)
processes. In the next level the BSS processes
are decomposed into order to cash process.
Further the order to cash process is divided
into order capturing process, order fulfllment
process and billing process. Typically, during
transformation, order capture is implemented in
CRM systems, order fulfllment is implemented
Figure 1: Example Of Process Levels In A Telecom Company Source: Infosys Research
OSS Process
B
A
M
L0-L1
Processing
L2-L3
Processing
L4-L..n
Processing
BPMS
BPMS/ERP BPMS/ERP BPMS/CRM
Billing Process Order Fulfillment Order Capture Process
Order to Cash Process
BSS Process
39
using telecom order management systems
(TOMS) and billing is implemented in telecom
billing system (TBS). Instead of a standard CRM
or ERP systems, these capabilities can also be
implemented using BPM platform.
So EA needs to determine on how
the best of these systems can be utilized to
ensure automation and agility to the system.
If the system has the capability to provide the
implementation of a business process with
less customization and the system provides
industry specific solution then building a
new BPM solution will require higher effort
for customization and longer implementation
cycle. So these capabilities must be utilized.
Wherever the ERP solution system does provide
an out of box support for business process
implementations the BPM products should
not be used. Typically, granular level business
processes (L3, L4Ln) are implemented in these
systems. These granular business processes
should be implemented using BPM only if
the systems do not support out of the box
customizations.
In a different context, if there are more
coherent higher level business processes that
require orchestration of granular business
processes implemented in incoherent multiple
systems then it is benefcial to have the higher
level business processes implemented in BPM
platform. Apparently, the implementation of
these business processes requires integration
with various systems and hence would involve
higher effort if implemented in one or more of
the packaged applications.
BPM is not limited to process management
or process modeling. The platform provides the
best capability for business activity monitoring
(BAM) and it makes good sense to use this BAM
capability. In EA blueprint the BPM integrates
with all the packaged applications thus enabling
the business users and managers to optimize
the business process according to their business
needs.
INTEGRATED BUSINESS COMPOSITE
APPLICATIONS
Business transformation impacts the entire
organizations boundaries. Traditionally an
organization is divided into business units,
domains and departments. To be agile and
flexible the business needs to be globally
i nt egrat ed and cannot operat e i n si l os.
Todays organizations need to break global
boundaries by bringing in new partners
t o i mprove product /servi ce capabi l i t y,
reduce cost and enter new markets. With
transformation every organization aims to
break business silos and create an integrated
business. Integration means more business
processes and applications will leverage the
capability provided by others both within
and outside organizational boundaries. In the
past, applications were typically purchased
or built in-house based on the business silos
requirements. With transformation, the goal is
to create composite services that leverage and
integrate services available from all the units
in the enterprise.
Compos i t e appl i cat i ons r equi r e
orchestrations of multiple business processes
implemented in various applications. The
business processes implemented in packaged
applications, BPM or using any other IT system
are exposed as service. Composite applications
are essentially course grained business processes
or services. They are implemented orchestrating
these services. BPM platform along with
SOA provides the capability to integrate
with enterprise information systems. These
composite applications should be implemented
in BPM platform.
40
Sometimes a composite application can
be created by the service where the underlying
services are implemented using the same
package or IT system. In this scenario also it is
recommended to create composite applications
in BPM platform for future extensibility. As
mentioned before, a thumb rule is to use
BPM platform to conceptualize the business
processes wherever possible. Then use SOA
as the tool to orchestrate and implement these
business processes as services.
Wh i l e c o n c e i v i n g b u s i n e s s
transformation and creating an EA blueprint,
it should be noted that one of the pillars
supporting a SOA is the definition of service
oriented business or business processes as
a service. The business process needs to be
defined considering each business process as
a business function and thereby a business
service. Secondly, each business process should
be exposed as service. Composite applications
will consume these services in order to reuse
the business process.
The EA blueprint thus will have a service
catalogue which is basically a derivative of
business process. Thus, SOA is the enabler for
service oriented business processes. It should
also be noted that companies must determine
which business processes or integration efforts
are well suited for SOA instead of putting each
and every thing in the SOA bucket.
An EA blueprint should also lay the
technology footprint. From this point of view the
BPM platform and the ESB must be positioned
together. Though a BPM tool can have the
capability to integrate with various enterprise
Operational
Service
Framework
Error
Handling
Auditing
Logging
Alerting
Security
Service
Business Partner Integration Framework
B2B Data
Exchange
B2B Partner
Management
Enterprise BPM / BAM Framework
Business Process
Management
Business Rules
Management
Business
Optimization
Enterprise User Interaction Framework
Content Management Personalization Integration People Collaboration Search
Network Solution Storage Solution DR Solution Server virtualization Database Solution
Enterprise Core Infrastructure Framework
Enterprise Business Application Framework
ERP
Core Business
Application
Channel
Application
Enterprise Business Information Framework
Data
Warehouse
Data
Marts
MDM
Enterprise Integration Services
Communication Application Access Data Service Service Wrapping Event Management
Business Services Delivery Framework
Service Management Service Access Service Composition
Figure 2: A BPM-SOA Reference Model Source: Infosys Research
41
i nformati on systems yet the i ntegrati on
capability of BPM must be used to integrate
with ESB. Figure 2 is one of the reference models
that lays out the various components in an EA
at a high level. SOA thus should be used as the
technical architecture concept in view of BPM in
an enterprise. Further, prioritization of services
or the business processes and functions help in
creating a robust EA and can be progressively
implemented and transformed into the end state.
As ment i oned ear l i er , when an
organization engages in a transformation
program, multiple system integrators are
involved. These systems integrators bring
with them their own frameworks for business
process implementations. While we have
already suggested how to go about using
BPM and SOA in the EA blueprint, it is
also advised that a top down approach be
employed to govern these implementations in
a transformation program.
Essent i al l y i n t he t ransf ormat i on
program, during the EA blueprinting phase,
business process catalogs and service catalogs
need to be defned. During the implementation
phase, planning of these process and service
implementations are fnalized aligning with the
overall transformation plan.
While cataloging or identifying these
business processes for the new business, they
are determined from the as-is analysis of
existing business process and the new business
requirement [Fig 3].
S u b s e q u e n t l y , i n t h e s e r v i c e
decomposition process, processes must be
mapped to application with respect to (a)
transactional cardinality, i.e., multiple intra/
inter application interactions are required
for the business process and transaction
volume; and (b) business process information
with application data objects and for each
integration point the data objects need to be
identified.
Based on the business process and the application
capability, the services will be identifed that
matches the business requirement [Figure 4].
As Is
Analysis
of business
process
Business
Driver
Identification
Domain
Based Process
Decomposition
Process
Catalogues
Process
Application
Maps
Figure 3: Determining Business Processes Source: Infosys Research
Figure 4: Determining Services Source: Infosys Research
Business Process
Catalogues
Application
Portfolio Analysis
Service
Analysis &
Decomposition
Service
Granularity
Business
Process to
Service Map
Service
Application
Map
42
Investment Review
Baseline Enterprise Service Registry
Service Functional Architecture Approval
Service Approval
Service Identification Service Review
Service Profile Service Profile Approval Service Profile Review
Service Functional Architecture Review Service Functional Architecture Baseline Service Profile
Trigger
Undergoes
Undergoes
Undergoes
If Complaint leads to
Form basis for
Drives
D
r
i
v
e
s
If Complaint
leads to
Form basis for
Form basis
for
The next step is that of identifying service
granularity. This is an important step for a
successful SOA implementation. There is
always a trade-off between cost of creating a
new service and using the existing service. In
this stage, two important factors business
alignment and ripple effect - are to be
considered. This will likely allow existing
services to be swapped with new service
implementations from different providers
or with newer versions without disturbing
business process functionality.
During the implementation phase, the
business requirements are determined. Based
on the requirements, business process and
services are identifed from the process and
service catalogs. Services and the processes
maps with the applications will help in
determining the system requirements of the
various systems including middleware.
Fi gure 5 depi ct s t he archi t ect ure
governance process for managing the service
catalogs in the enterprise. This well-defned
process of creating a service application map
and carrying forward the implementation
ensures implicit governance through the
process to establish a SOA layer in the EA.
Since business transformations is progressive
in nature and till the end state is reached new
systems and processes will likely sit alongside
legacy systems and processes requiring systems
integration implementation. In the end these
legacy systems will be replaced or merged
with the new business processes as per the EA
blueprint and implementation plan. During the
transformation the new business processes has
to be supported by both new systems and the
legacy systems.
Needless to say, it requires a series
of business-evolutions to achieve the end
goal of a transformati on program. SOA
emphasizes on the evolution rather than big
bang implementation. SOA abstracts the
business service implementation details from
Source: Infosys Research Figure 5: Business Service Catalog Governance Process
43
the service consumers. In the transformation,
service oriented approach abstracts the new
and legacy systems to implement the business
service required for the new business process.
Service implementations are abstracted from the
consumers so that the new systems can replace
the legacy systems in a phased manner without
impacting the new business process.
CONCLUSION
Business organizations need to grow beyond
their current state, improve effectiveness in
BPM, improve on their IT services and make IT
infrastructure scalable for non-linear growth.
They need to be agile and adept in changing
to the needs of time. It is therefore crucial to
achieve business transformations on the wheels
of SOA and BPM. By effectively employing the
tools of SOA and BPM in systems integration
helps formulating a blueprint for EA in
achieving business transformation objectives.
In a business transformation there is always
an affinity towards disjointed implementation
of best of breed products causing a rigid
enterprise post transformation. More emphasis
must therefore be given to EA governance
in order to achieve the business goal set
for transformation. Without well-defined
governance process there is a risk of failure in
any business transformation program.
REFERENCES
1. Durvasula, S., Guttmann, M., Kumar,
A., Lamb, J., Mitchell, T., Burc, O., et al.
(2008). SOA Practioners Guide. Retrieved
from SOA Blueprint (Strategic Use of IT).
Available at http://www.soablueprint.
com/yahoo_site_admin/assets/docs/
SOAPGPart1.290211145.pdf.
2. Heal t h, N. I . ( 2007) . Workf l ow/
Bus i ne s s Pr oc e s s Ma na ge me nt
(BPM) . Retrieved from Enterprise
Ar chi t ect ur e NI H. Avai l abl e at
ht t p: //e nt e r pr i s e a r c hi t e c t ur e .
n i h . g o v / A r c h L i b / A T / T A /
WorkfowServicePattern.htm.
3. Anonymous.(2007). Analyze the business
and define the target environment.
Retrieved from US Department of the
Interior. Available at http://www.doi.
gov/ocio/architecture/mbt/step3.htm.
4. Channabasavaih, K. , Holley, K. , &
Tuggl e, E. ( 2003) . Mi grat i ng t o a
Service Oriented Architecture Part 1.
Retrieved from IBM Developerworks.
Ava i l a bl e a t ht t p: //www. i bm.
com/devel operworks/l i brary/ws-
migratesoa/.
5. Koch, C. (2005). A new blueprint for the
enterprise. Available at http://www.cio.
com.au/article/30960/new_blueprint_
enterprise.
6. Mal hot r a, Y. , ( 1996) . Ent er pr i s e
Architecture: An overview. Available at
http://www.kmbook.com/entarch.htm.
7. Kulkarni, N., Dwivedi, V. (2008). IEEE
SCC 2008. Retrieved from IEEE: The Role
of Service Granularity in a Successful
SOA Realization A Case Study.
8. Padmanabhuni, S., Ganesh, J., & Moitr, D.
(2004). Web Services, Grid Computing,
and Business Process Management:
Expl oi t i ng compl ement ari t i es f or
business agility IEEE International
Conference on Web Services.
9. Behara, G., Mahajani, P. and Palli, P.(2010),
Telecom Reference Architecture, Part
2, www.bptrends.com. Available at
http://www.bptrends.com/476009B9-
D3DC- 4686- 9969- 1F747296B3A5/
FinalDownload/DownloadId-BB17D4
B33E56270161F7CB477CADD9D1/4760
44
09B9- D3DC-4686-9969-1F747296B3A5/
publicationfles/EIGHT09-14-10-ART-
TelecomReferenceArchitecture-Part1-
Beharaetal.pdf.
10. Anonymous (2011), Infosys Telecom Order
Management Solution Available at http://www.
infosys.com/CRM/offerings/Documents/
telecom-order-management-solution.pdf.
45
VOL 10 NO 2
2012
Adopt an ESB framework and mitigate
risks inherent in your application
integration programs
By Prakash Rajbhoj and Narayanan Chathanur
A
pharmaceutical company has its customer
data maintained in a CRM system. The
company merges with a healthcare system
and the combined entity needs to integrate
its customer base. Can enterprise application
integration (EAI) help in this process?
A texti l e company has i ts suppl y
operations including raw material delivery
from its vendors to end product delivered to
various stores through different transportation
systems such as trucks, trains and cargo planes.
Its inventory management as well as transport
is controlled with AS400 systems. The company
decides to upgrade and get new ERP systems
to optimize the whole process with an objective
to improve effciency by 30%. The ERP system
now needs to be integrated with other modules
in AS400 and some other discrete applications.
Centralized services need to be created as part
of this initiative. Is it case for EAI or service
oriented architecture (SOA)?
A banking institution has a large number
of B2B transactions and large fles are to be
settled with corporate clients. The process is
time consuming and is crying for an upgrade.
Can SOA i f empl oyed i mprove i ts ti me
effciency?
An insurance company is facing the risk of
a waning customer base owning to its competitors
that are offering similar services at cheaper rates.
This is coercing the company to offer attractive
customized products to its customers even while
keeping an eye on its operational costs. Such an
effort calls for incredibly fexible IT systems. Can
SOA be of any help?
A manufacturing company bought an
EAI product from an established vendor to
integrate its settlement system with different
business lines. There are similar other projects
that are in progress that use the same EAI
tool. But the vendor increased the license
costs and the company has little choice as it is
dependent on the EAI tool for its integration
needs. Can vendor product be replaced by
adopting SOA and providing more open
standard solution?
Infosys Labs Briefings
Applications Integration with ESB
46
Most of the above are typical cases that
transgress organizations be they small scale or
large scale and IT needs to adopt newer ways
to help businesses achieve their objectives. EAI
and SOA exactly do the same by offering levers
in systems integration space. In this paper, we
attempt at feshing out the value proposition of
EAI and SOA, their synergy and the essential
ingredients for their successful implementation.
EAI AND ESB
IT systems upgrade can be classifed as:
Buying out end-to-end homogeneous
ERP packages replacing almost all
the existing IT software systems and
aligning the business processes with the
ERP recommended processes.
Procuring standalone commercial off-the
shelf (COTS) applications in identifed
segments to i mprove the exi sti ng
businesses processes.
Enhancing the current obsolete systems
with open standards based system.
Most Fortune 500 companies have
mainframe or AS400 systems that are difficult
to replace as they already perform a lot of
complex transactions. These though still need
to be integrated with other IT systems thereby
necessitating the creation of wrappers around
different modules of the legacy systems to
align with the modern systems. These are
typical cases where there is an obvious need for
EAI. Figure 1 illustrates on how an EAI pattern
of hub and spoke model is used to integrate
different applications. EAI helps address
problems that arise out of point-to-point
integrations like for example lack of stability,
extensibility and inadequate interoperability
framework. However, EAI implementation is
not devoid of its share of inadequacies. First, its
contribution to improving the actual business
process is questionable as it fails to plug any
gaps in the business process layer. Second,
EAI implementations are not only complex
but also cost intensive and require long term
commitments from the middle vendor which
may in turn lead to vendor lock in related issues.
SOA, with its ability to fully abstract
proprietary technology, to a large extent
mitigates the risks incumbent in EAI. Vendor ties
can be broken more easily and the organization
implementing an upgrade can gain more control
over its integration architecture.
Several innovations that were popular
during the EAI era were reckoned as useful to
the overall goals associated with building SOA.
Different Systems in an
IT Landscape of an Organization
ERP
System 1
Legacy
System 1
ERP
System 2
Adapter Adapter Adapter
EAI Hub (Messaging Service)
Adapters offered by Middleware
vendor to connect to various
applications to a common ground
Middleware Features
Orchestration Engine
(Running Different
Processes Cohesively)
EAI
Vendor Proprietary
MiddlewareTechnology
Complex Due to
multiple integration
transformation rules
Broker (System
to System Message
Transformations)
Figure 1: A Simple Illustration of EAI
Source: Infosys Research
47
One such example is the broker component
that al l ows for servi ces usi ng di fferent
schemas representing the same type of data to
communicate through runtime transformation.
The other is the orchestration engine that can be
positioned to represent an entire service layer
within larger SOA implementations.
With the advent of SOA, services became
the focal point in the enterprise architecture. In
an ideal world service producers and service
consumers are expected to communicate
with each other without an intermediary but
unfortunately the reality is quite different. Most
services are built out of existing applications and
had proprietary data structures and protocols
that inhibited communication between the
producers and consumers. ESB has filled in
this gap as an intermediary and evolved as a
communication nerve center for such services.
ESB provides a robust middleware infrastructure
for connecting application and services.
ESB itself is based on a messaging
platform generically called as an enterprise
mes s agi ng s ys t em ( EMS) t hat al l ows
communication through open standards such
as XML, simple operation access protocol
(SOAP) and web services. EMS inherently
should take care of standard requirements such
as security, routing, metadata, subscription
and policy which the vanilla messaging layer
of EAI fails to address. Further ESB calls for
an abstract layer over the EMS where base
functions are built upon and then broken
into constituent parts and deployed in a
distributed architecture method. Thus the
unintelligent hub of EAI now translates to
multiple intelligent hubs though serving a
common purpose. ESB also allows various
middleware vendors to create their product
sui te that supports open standards and
interoperability with other applications easily.
The various advantages thus ESB-based EAI has
over traditional EAI are:
ESB provides standards based, fexible,
and consistent architecture for your
integration.
ESB r educes t i me and ef f or t t o
create new processes through reuse
of exi sti ng appl i cati ons and data.
ESB is an event-driven and message
oriented platform that promotes SOA.
ESB facilitates uniform application of
business rules, enrichment of messages
from other sources, spl i tti ng and
combining of multiple messages and
common exception handling.
ESB increases flexibility to change
complex system behavior by minimizing
t he hi dden dependenci es among
applications and services.
ESB supports service management, for
e.g., virtualization and substitution of
services, service metering and policy
management, etc.

ESB has thus emerged as the one of the
bridging blocks for EAI and SOA. Adopting
ESB however does not necessarily mean that
organizations have become self-sufficient and
are able to build their middleware in line
with SOA. ESB facilitates SOA but adopting
vendor-based ESB technology in itself is not a
panacea for all integration challenges. There
are many implementations where expensive
technology like ESB has been used or should
we say misused without any tangible ROI. To
48
address such challenges, a SOA based reference
architecture is needed for implementations.
Most of the cases that were articulated in the
introduction this paper can be solved with this
SOA based EAI implementation.
SOA-BASED REFERENCE ARCHITECTURE
FOR EAI IMPLEMENTATION
Figure 2 depicts a reference architecture
showcasing the multi-tier SOA setup for large
implementations. This architecture takes care of
Source: Infosys Research
Reference SOA Architecture for EAI Implementation
Security
Infrastructure
KPI based
business activity
monitoring
Error Handling
Framework
Automated
Testing Framework
Auditing
Framework
Alerting
Framework
Directory
Service
Logging
Framework
Operational
Service
Framework
Service Management Layer
Service Manager
Auditing Monitoring Registry Routing
Security Versioning
Business Process Layer (People, Integration, Decision, Content)
Modeling Execution Simulation Analysis Business Rules Content Integration
Process Management Layer
Integrated Services View Portal
Access Controls
Authorization & Authentication
Reports
Content Delivery
Customer Setup
User Management
Presentation Layer
Service Delivery
WSDL Wrapping Aggregation Orchestration Business Rules Exception Management
Service Delivery Layer
MDM Data Layer
Business Information
Data Storage Data Semantics Data Caching
Data Federation Data Rules
Data Management Workflow
Data Layer
Adapter
Organization IT applications across different business units that need to be integrated
Message Broker
Enterprise Integration Layer
Message Layer
Transformation Validation Routing Enrichment Data Correlation
Process Automation Data Integration
App 1 App 2 App 3 App 5 App 6 App 7
Mutiple App 4
versions
across units
Figure 2: Reference Architecture Based on SOA for EAI
49
different aspects of integration such as process,
data, business, and applications.
As seen in the fgure, the applications
and software systems are integrated through
a messaging layer in the form of adapters.
Any service or application that wants to
communicate with other application will
have to go through this layer. This provides
for decoupling and componentized view of
applications.
Any new application that needs to be
integrated will just need to connect through this
adapter and messaging layer. The messaging
layer is the interface between all communication
layers such as adapter, broker, service delivery
and interconnection to process management
and user interface (UI). It could be a hub like
JMS server as noted before or it could be a
protocol such as HTTP or any way of open
standard communication.
Service delivery conforms to SOA
standards and is a logical representation of how
the messages should be formed complying with
standards such as Web Services Description
Language (WSDL), Representati ve State
Transfer (REST), etc. It also provides the
common functions layer as part of ESB such as
exception handler, data caching framework etc.
Service management layer on the other hand
takes care of SOA governance mechanisms such
as service versioning and security.
Process management is a logical layer
that deals with business process modeling to
business rules concepts created that basically
holds the code for integration transformations
and orchestrations. Presentation layer is a
single user interface portal that caters to
all requirements related to integration. It
fetches the details through the service layer
which in turn is connected through different
applications.
Continuous improvement is one of the
goals of SOA and one of the ways is to establish
patterns that can be reused as a standard
implementation.
ESTABLISHMENT OF PATTERNS IN SOA
FOR EAI
In EAI, with ESB as the underlying architecture
pattern, a basic set of patterns are highlighted
in Figure 3.
These patterns can be used based on
the scale of EAI implementation. Also another
factor is that of considering deployment
pattern once the services in ESB as a core as
well as the various layers are developed. As
seen in Figure 4 ESB combines both the bus
architecture advantage as well as the hub
architecture.
Further during the creation of ESB,
following components of the ESB can be
created with each of them serving a set of
capabilities.
Communication: Thi s part of ESB
should have the capability of facilitating
rout i ng, addressi ng and prot ocol
support (http, https), publish/subscribe
request/response, fre/forget events,
synchronous and asynchronous events
and messaging.
Quality of Service: Facilitate transaction
management (atomi c transacti ons,
compensation or WS-transactions) and
assured delivery paradigms indicating
reliable messaging.
Service Level: For non-functional needs
such as performance, throughput,
availability and basic for contracts and
agreements.
50
Message Pr ocessi ng: Encodi ng/
decodi ng, dat a t r ansf or mat i ons,
correlation, aggregation, storing and
forwarding are some capabilities that
are required.
Modeling: Obj ect model i ng, dat a
formats and other data constructs and
structures that form the messaging in
the overall enterprise.
Monitoring: Policy driven pattern
recognition and dynamic business rules
management for providing activity
monitoring support.
SOA implementation is not complete without
sophisticated SOA governance that takes care
of automated mechanism as well as manual
oversight using reports and analysis of data on
a periodical basis.
HOW BEST TO LEVERAGE SOA
GOVERNANCE
ESB to a large extent is a bottoms-up approach
for integrating and composing business services
in the organization. This approach focuses on
integrating services around individual ESBs
that can be quickly assembled. Whatever be the
approach for SOA, one still needs the capability
to track, manage and control the usage of
services in ESB. This primary governance
aspect is divided into two types -- design time
governance and runtime governance
Design time governance as the name
suggests implies typically providing integrated
service registry/repository that attempts
to manage a service from its design to its
deployment but not during its runtime. This
necessitates ESBs to provide such registry/
repository internally or it should be easily
integrated with third party service registry/
repository. Universal Description Discovery
Direct
Brokered
Federated
Multiple namespaces, administration
domains; namespace mapping in
each ESB; services are likely to be
applicable throught the entrprise
Multiple namespaces, administration
domains; namespace mapping in
Federated ESB facilitates services
interaction with multiple
implementations; subset of services
applicable throughout the enterprise
Multiple namespaces, administration
domains; namespace mapping in
Brokered ESB Gateway facilitates
services interaction; subset of
services applicable throughtout the
enterprise
Figure 3: : ESB Integration Patterns Source: Infosys Research
51
and Integration (UDDI) based registries are
very popular for design time governance
and have been supported by almost all ESB
products.
Runtimes governance is essentially
used on already deployed services to enforce
policies and service level agreements (SLAs)
based on the need for compliance, security
and performance. There are no de-facto
standards for runtime governance but certain
patterns like security, service validation,
logging and auditing, setting and maintaining
appropriate service evels and managing errors
and exceptions have emerged with respect to
services hosted in ESBs.
Typically ESBs have runtime policy
management tools that are used to apply and
enforce policies on runtime. This ensures that
implementation teams can concentrate on
building functionality in the service, leaving
the runtime policies to be confgured at a later
stage when service is deployed. Also it provides
flexibility in changing the policies without
impacting the service behavior.
PRODUCT VENDORS SUPPORT TO ESB
With the predominance of SOA as an architecture
model, many EAI vendors started supporting
ESB functionality in their EAI product suite.
XML has become the de-facto standard for all
messaging and EAI vendors added support
for web services in their existing tool set. Some
vendors have repackaged their EAI tool as ESB
to ride the bandwagon of SOA. Following are
some of the changes seen in the EAI product
suite of major vendors
Support for SOAP as a messaging model
and XML for carrying the message
payload.
Figure 4: Deployment of ESB Compared to EAI Source: Infosys Research
Enterprise
Service Bus
Configuration
Control Client
<HUB>
Runtime Node
<HUB>
Runtime Node
<HUB>
Runtime Node
<HUB>
Runtime Node
Service
client
Service
client
Service
client
Service
Provider
Service
Provider
Service
Provider
Configuration and Control Services
Distributed Infrastructure Single Point of Control
Service
Client
Service
Client
Service
Client
Service
Client
Service
Client
Service
Client
<HUB>
ESB
Service
Provider
Service
Provider
Service
Provider
Service
Provider
Service
Provider
Service
Provider
<BUS> ESB
52
Web Services wrapper for application
and technology adapters
Support for WS-* standards for the web
services.
Servi ce Component Archi t ect ure
(SCA) support so that the integration
components can be easily exposed as
services and can be bound to multiple
protocols.
Orchestration engines that can create
business processes around the already
existing services.
SOA governance framework and service
registry.
Governance aspect as well as applying
SOA on EAI could have been diffcult without
out-of-the-box and integrated tools support
from EAI vendors. Traditional EAI vendors such
as TIBCO, IBM, Oracle, and SAG have upgraded
their EAI product tool set to provide out-of-the-
box support for SOA standards such as JMS,
SCA, web services and SOAP. The product set of
these vendors provides support for each of the
layers in the reference architecture as indicated
in Figure 2. Thus keeping up with the SOA
and support to open standards these products
can easily work interchangeably with other
ERP tools or custom tools complying to the
standards. For e.g., TIBCO Active Matrix Suite
is fully compliant and ready to go from design,
development and post production governance.
Such products also provide off-the-shelf
programs that would help in creating a set
of services and frameworks to create an ESB
to serve as the integration platform for any
enterprise.
CONCLUSION
All types of enterprises whether small or large
always have the need for EAI. Traditional EAI is
expensive and has its own incumbent problems.
SOA offers the much needed relief by plugging the
problems inherent in EAI. ESB is one such concept
that binds both SOA and EAI. As sketched out with
this paper, enterprises can use these concepts to
create the integration platform for their IT systems
by using the pillars of EAI and SOA and solve their
typical and non-typical business problems.
REFERENCES
1. Goel, A.(2006), Enterprise Integration
EAI Vs SOA Vs ESB, Avai l abl e at
h t t p : / / g g a t z . c o m/ i ma g e s /
E n t e r p r i s e _ 2 0 I n t e g r a t i o n _ 2 0 -
_20SOA_20vs_20EAI_20vs_20ESB.pdf.
2. Sarkar, M. (2011), Moving from EAI to
SOA- Infosys Perspective. Available at
http://www.infosys.com/consulting/
packaged-application-services/white-
papers/Documents/eai-soa.pdf.
3. Twite, B.(2007), EAI to SOA: Evolution
or Revolution. Available at http://www.
marlo.com.au/pdf/EAI_to_SOA%20.
pdf.
4. Berg, J. L. (2011), The Integration Between
EAI and SOA. Available at http://www.
soamag.com/I49/0411-2.php.
5. Earls, A. (2011), How EAI and SOA
play complementary roles in the SAP
ecosystem. Available at http://www.
nagarro.com/Documents/Search_SOA.
pdf.
6. Pentakalos, O. (2006), Introduction to Web
Services, SOA, and ESBs- An enterprise
integration persective.Available athttp://
www.sysnetint.com/presentations/
nterpriseIntegrateionWithSOAAndESB.
pdf.
53
VOL 10 NO 2
2012
Enterprise solution architecture provides an end-
to-end view of the entire solution being developed
in all phases of the transformation program
Solution Architecture in Large
Transformation Programs
By Suchit Bhatwadekar
Infosys Labs Briefings
I
t is common knowledge that huge investments
are sunk into transformation programs with
the expectation of highly favorable business
outcomes. Such programs shake the central
nervous system of the enterprises undertaking
them. Transformation programs are fraught
with uncertainties and unforeseen risks. A
good solution architecture to a large extent can
mitigate such risks.
At present, there are various architecture
perspectives that are in practice. While enterprise
architecture (EA) frameworks revolve around
strategy and vision aspects at enterprise level
and fail to focus on transformation programs,
service oriented architecture (SOA), business
process management (BPM) and model driven
architecture (MDA) are not able to go beyond
the realm of web services and integration. In
nutshell none of the architecture methodologies
seem to suffce the needs of solution architecture
in a transformation program.
Ent er pr i se sol ut i on ar chi t ect ur e
(ESA) that is being considered the next big
thing in EA space is slowly inching towards
mainstream adoption and is likely to become
a force to reckon with in the near future.
The fundamental difference between other
architectures and solution architecture in
transformation programs is derived from their
scope, boundaries and success criteria matrix.
Solution architecture defnes end-to-end
view of the entire solution being developed in
all phases of the transformation program. It
builds the traceability between business drivers
and the product/process capabilities deployed
along with performance metrics for continuous
improvement.
On scope, solution architecture is not
just limited to deploying new product but also
takes care of desperate as-is landscape of legacy
to re-aligning the end-to-end business process.
Similarly, scope goes beyond one technology
54
as well cuts across multi-country federated
environment.
Boundaries of solution architecture
touch every aspect of the enterprise i. e. ,
design for stringent service level agreements
(SLAs), consumption of fewer resources, easy
maintenance and optimization of every line of
code to deliver the best performance despite
increase in business volumes.
Solution architecture will evolve over the
lifetime of a program and should (a) identify
the expected nature of the solution and risk
involved; (b) be revised in line with increased
understanding of the requirements and the
various options available to deliver the solution;
and (c) continually refect the end-to-end view
of the solution being implemented.
The key deliverables of solution architecture are:
Executable business blueprint that
maps the end-to-end business process,
capabilities and existing applications for
functional completeness.
Adaptable technology blueprint that
ensures that the hardware and software
architecture match the scalability,
security and robustness of solution to
ensure technical compliance.
Integration architecture to manage
seamless integration during and after
transition for consistent integration.
DRIVING PRINCIPLES OF SOLUTION
ARCHITECTURE
Unlike any other architecture paradigm, solution
architecture works on absolute data points.
For example, having state-of-art-web service
implementation might provide long term
benefts, but not an excuse for nonconformance to
response time SLA at an ATM. If there is a need
for P2P integration or change in the core product
to meet the SLA, then so be it. It is important for
solution architecture to prioritize hygiene factors
of running the business without any casualties.
SLA dri ves t he core of sol ut i on
architecture. SLA can be business and/or
technology driven. For example response time,
transaction integrity, availability, disaster
recovery, financial impact, and time-to-
market, response to customer or monitoring
and traceability etc. Solution architecture
goes beyond 99.99 parameters to ensure the
executability of a deployed solution.
Solution architecture builds traceability
from change to changed while traversing
through the transformation elements. To
approach this paradigm it is important to adopt
a precise sequencing method with concomitant
quality metrics as part of solution architecture.
Such a method could have multiple views of
solution architecture, viz., business view, co-
existence view, integration view, SLA view, risk
view, and technology view.
Based on a change, business view
captures each to-be process; coexistence view
validates the realignment of new process to the
legacy application processes; integration view
captures the functional and technical integration
specifcation of that process to realize end-to-end
business fow; SLA view intercepts the business
SLA to technology SLA; risk view validates the
exception and critical case; and technology view
builds the technology infrastructure to meet the
business and technology SLAs.
INTERPRETING BUSINESS IN
TRANSFORMATION PROGRAM
Integrated view of business change is the frst
critical success factor for transformation. It sets
55
a clear investment direction, business beneft
guidelines and coordinated people-process
change management. This needs enormous
deliberation on people and process aspects with
multiple stakeholders before conversion to the
next generation business blueprint.
Business View
The basic objective of business view is to map
the expected change to business i.e., operation
excellence, new product launch and its lifecycle,
core - support processes, etc. Key activities
involved are (a) mapping between target
operating model and transformation objective;
(b) alignment of transformation objective with
business capability model; and (c) deriving to-
be processes.
To start with, organizations assess the
impact of change in present business model
and redefine new business capabilities or
alter existing ones. These capabilities are then
mapped to the to-be processes. For example,
when a bank wants to enhance its distribution
capabilities that involves distribution of third
party product and services including that of
fnancial subsidiaries of the group, through
its existing channels or that of distribution of
banks products through third party channels
like retailer, real estate agent, etc., distributable
product factory is the new business capability
and multichannel architecture to distribute the
product is technical capability. On the process
side, new to-be process requires alignment
to existing CRM process to defne end to end
business fow.
Before finalizing the to-be processes,
it is crucial to assess the impact on existing
applications and processes. If the existing
application portfolio alteration (legacy changes)
cost is more than the transformation costs then
it negates the case for transformation.
For example, if the bank decides to
change its account opening process to improve
operation efficiency, it would require the
revamping of the complete channel application.
In a way if it calls for a huge retroftting in any
existing application (e.g. loan, investment, etc)
then change has to be revalidated with data
points like feasibility, cost, time and schedule
before adopting the new process.
Coexistence View
Coexistence evaluates integration with the
legacy system, especially of that existing
application that has not been replaced during
transformation. It also deals with managing
both the old and the new system during the roll
phase of transition.
Coexistence view realigns the to-be
process with existing application processes.
Realignment involves detailed analysis of
process distribution and their dependency,
business entity distribution, multiple channels
i nteracti ons, etc. Thi s exerci se bubbl es
out various techno-functional challenges.
Identifying the alternatives for each case needs
in-depth workshops with individual owners to
evaluate feasibility that ultimately drives the
applicability of change in the organization. This
view studies such applicability on the following
parameters.
C o e x i s t e n c e l o g i c ( r o u t i n g ,
reconciliations, etc).
Changes retroftting efforts in existing
application.
Strategy to minimize the changes.
Functional integration between new and
existing applications.
56
Business entity and information exchange
between new and existing applications.
Develop the blue print of integration
requirements.
Integration View
While coexistence view provides the blue print
for integration, the integration view hardens
these integration requirements into technical
frameworks to support the business change.
Integration view ensures transaction integrity,
guaranteed routing of message, monitoring
business exceptions and reconciliation to
resubmit, etc. , that is more than writing
transformation rules between two applications.
Huge analyses is required to understand
existing application orchestration, identify the
breaking points and structure the flow that has
less impact on business in to-be. Integration
view applies the following data points to arrive
at a consistent integration.
Integration SLA.
I nt egrat i on pri nci pl es t o reduce
efforts at existing application.
Integration patterns.
Common messaging formats.
Tools and framework defnitions (error
and exception handling, transformation
and orchestration).
Technical implementation.
INTERPRETING TECHNOLOGY IN
TRANSFORMATION PROGRAM
Managing constant change and ensuring it is
executable is the next critical success factor
for a transformation program. While business
blueprint encompasses the manufacturing
aspect of change, technology blueprint defnes
the design aspect of change. There are various
facets of technology blueprint like the extent
of adaptable design to support future changes,
the extent of infrastructure to be built to
run business as usual for present and future
requirements, how to minimize the impact of
new changes arising in future and the way in
which return on investment (ROI) is assessed
while purchasing new resources.

SLA view
The bas i c obj ect i ves of SLA vi ew i n
transformation program are doing a 360 degree
evaluation of SLAs, primarily to assess the
post-transformation impact on service levels;
aligning business SLA with technology SLA;
and deriving the present and future SLA. The
risk of technical SLA mismatch is inevitable
during the rip and replace of the application
in enterprise landscape. For example, while
the execution of as is process may happen in
milliseconds due to a homogeneous platform in
current applications in to be the process might
get distributed across old and new application
platforms. This is bound to effect the response
time of the SLA.
Apart from technical SLA, this view also
strategizes the adaptability of new solution to
future changes. For example, post transformation,
new channels are added or one of the existing
applications replaced by new ones. In these
situations integration specifcation should be
designed in such a way that the impact on both
sides is minimized to negligible. Similarly, by
adding extra infrastructure resources, new
platform should be able to absorb additional
volume of transactions and customers.
57
SLA vi ew desi gns the traceabi l i ty
mechanism to identify the bottleneck and
corrective procedure across all technical
components.

Risk View
Risk management is a de-facto in any program,
but in transformation program it shoots
beyond 99.99 parameters to ensure robustness
of the solution. The replacement of core
system comes with a set of risks. The solution
architecture planning should include a risk
management methodology and mitigation
plan to ensure the solution implementation
achieve the end objective without jeopardizing
customers reputation, business operations and
infrastructure.
Risk can impact client in terms of --
Poor customer service and overall
operations.
Implementation schedule and timelines
overrun.
Technology specifcations requirements
User requirements.
Default in regulatory compliance.
Additional costs.
Risk view gravely validates each SLA
with associated risk. In a few critical and
exception cases, risk mitigation need not
necessarily nullify all impact and hence there is a
need for skewed view to plan for alternatives or
take bold decision during solution architecture.
The postmortem analysis of disaster after go live
will not help much because the damage might
have reached an irreversible state. Risk view
therefore goes beyond the SLA to ensure the
robustness of the solution.
Technology View
While information blocks are built up properly,
baseline technical infrastructure also needs
scalable design accordingly. Technology
view fosters the foundational capabilities
to (a) manage scalable and fault tolerance
infrastructure design; (b) monitor and audit;
(c) engage in disaster recovery and business
continuity; (d) ensure backup and recovery;
and (e) manage security and network.
Technology view collates all the key
inputs from SLA, risk and integration view to
formalize the above strategy for each technical
component in solution. The organization also
expects to reuse current infrastructure to a
great extent possible. This could be another
constraint that needs detailed consideration
while designing the solution.
CONCLUSION
Solution architecture is a framework to develop
end-to-end view of change that fully suffces the
business needs, ensures technical compliance
within budget and has a maintainable schedule.
In a transformation program, success of any
solution is heavily dependent on the acceptance
of the solution by concerned stakeholders.
Hence to create the right perception one needs
to continuously engage the stakeholders
with the right data-points and a convincing
implementation approach. Transformation
related challenges are very dynamic in nature
and need continuous validation and measuring
mechanism. Alongside these views, solution
architecture methodology also plays a critical
role in transformation. The methodology
58
has to sprint the aforementioned views in an
executable manner and needs to apply more
rigorous approach to validate and measure the
SLA to deliver guaranteed outcome.
REFERENCES
1. Burton, B., Allega, A. (2011), Hype
Cycle for Enterprise Architecture, 2011.
Available at http://www.gartner.com/
id=1751916
2. Garl an, D. , Shaw, M. ( 1994) , An
Introduction to Solution Arhitecture.
Available at http://www.cs.cmu.edu/
afs/cs/proj ect/vit/ftp/pdf/intro_
softarch.pdf
3. Owe n, M. a nd Ra j , J . ( 2 0 0 3 ) ,
B P MN a n d B u s i n e s s P r o c e s s
Management : I nt roduct i on t o t he
New Busi ness Process Model i ng
Standard, Popkin Software. Available
at ht t p: //www. omg. or g/bpmn/
Documents/6AD5D16960.BPMN_and_
BPM.pdf
4. Lusk, S. Pal ey, S. and Spanyi , A.
(2005), The Evol uti on of Busi ness
Process Management as a Professional
Discipline, in BP Trends. Available at
http://www.bptrends.com/476009B9-
D3DC- 4686- 9969- 1F747296B3A5/
FinalDownload/DownloadId-37D84C5
DC0CCBEF26540CBE77F435E4D/4760
09B9-D3DC-4686-9969-1F747296B3A5/
publ i cat i onf i l es/06- 05%20WP%20
ABPMP%20Activities%20-%20Lusk%20
et%20al2.pdf
5. Smith, H., Douglas, N., Ferrara, L.. Hayden,
F. (2002).The Emergence of Business
Process Management, A Report by
CSCs Research Services. Available at
http://alarcos.inf-cr.uclm.es/476009B9-
D3DC- 4686- 9969- 1F747296B3A5/
FinalDownload/DownloadId-FCA53C
AD73A5A038521D0F9B7319B837/4760
09B9-D3DC-4686-9969-1F747296B3A5/
doc/ps gc/doc/l ec/par t e4b/cs c-
emergenceBPM.pdf.
59
VOL 10 NO 2
2012
Ride on to the co-creation wave with able help
from your systems integrator
Role of Systems Integrators in the
Co-creation Process
By N.Vijaykumar
Infosys Labs Briefings
I
nnovate is just not an eight letter word.
It reflects a transformation in thinking
and acting. Innovation is no more seen as
a tactic to beat market downturns, but as a
continuous business strategy. It has become
a growth engine for organizations, to churn
new products and services and to improve
productivity. Value creation has become one
of the key drivers and most organizations -
big or small, include innovation as one of the
important business tenets to achieve proftable
growth, customer satisfaction and to get the
much needed competitive edge. Innovation
needs massive drive from the organization and
more importantly, a calibrated discipline to
execute and ensure success of such initiatives.
Information technology (IT) has emerged as an
enabler for driving innovation.
Innovation is no more restricted to the
inner confnes of an organization. It has begun
to spread to a wider eco-system and suppliers,
service providers, customers and even society
have become part of this journey. For a long time,
the customer-supplier relationship was restricted
only to transactions. However, in todays fercely
competitive world businesses have realized that
this relationship has to be taken beyond just
transactional levels to derive mutual business
benefits. And, in a quest to nurture their
relationships, customers and service providers
are now getting on to a common ground, helping
each other mutually and to innovating in the way
in which services are provided. This emerging
behavior, called co-creation helps service
providers understand clients businesses and
tailor their services accordingly, while customers
gets a much better deal value for money and
most importantly, the joy of creating what
they wanted. Also co-creation helps building
sustainable and repeatable business models.
Product, technology vendors and other
consulting frms are increasingly collaborating
with organizations to drive co-creation. System
integrators (SIs) who help organizations set
up the IT platform, leverage their knowledge
of both business and IT domain to create and
60
customize solutions to specific set of customers,
industry verticals and geographies. However
it is easier said than done. The eco-system
thus created is difficult to manage. There
are herculean challenges coming in the way
of partnering with a customer to co-create.
The success of co-creation lies in managing
this joint venture and ensuring that focus is
always towards the common goal. Though
co-creation is not limited to IT industry, for
the purpose of this paper, the viewpoints are
more focused towards IT and SIs refers to IT
service providers.
WHAT IS CO-CREATION?
Co-creation can be defned as an activity of
collective creativity, developing products
and services by collaborative execution and
based on mutual trust, thereby creating and
enhancing value for all stakeholders concerned.
Co-creation represents a personalization or
customization of a broader concept to suit to
a specifc or a set of requirements and creates
a platform for newer forms of interaction,
sharing, collaboration, thereby creating value.
The term co-creation was frst introduced by
Prahalad and Ramaswamy in their article
Co-Opting Customer Competence [1].
Figure 1 below indicates the co-creation
ecosystem that fosters ideation to growth.
At the core of this system is the co-creation
continuum that provides a central platform
for interaction. Customers, suppliers and end
users come together on this platform to engage
in creating mutually beneficial products and
services.
The growth of technology, particularly
proliferation of internet and social networking
has fuelled the growth of innovation within
organizations. Various innovation methods like
open sourcing and crowd sourcing amongst
others help the organization to source ideas
and contributions from various stakeholders.
Consumers become supreme and their views
are listened better than ever.
There are multiple forms and fashions
of co-creation that can be achieved between
various stakeholders, for e.g., co-creation
between organizations and the customers whom
they serve; co-creation between organizations
and their partners; innovation within the
company; between different business units.
For sake of focus, this paper restricts itself to
addressing co-creation between a supplier and
a customer organization.
In the IT industry, there has been
an i ncreasi ng t rend where t echnol ogy,
product service providers and SIs have been
collaborating with each other to develop unique
products or services that is likely to help thee
customer organization to build either stable
technology platforms to support their business
growth or build productivity improvement
tools and frameworks that will make their
operations effcient and productive. Either ways
the crux is to create value.
CO-CREATION AND IT
IT has been a maj or enabl er i n dri vi ng
innovation in organizations and has helped
to leverage the technological advancements to
derive maximum benefits. Though innovation
and co-creation have been existing for long
in manufacturing sector, innovation in and
using IT shot into prominence in the early 90s
when open source was considered as the next
big bet of that time. Since then, both customer
organizations and SIs have always been
using IT to bring in the required differential
advantage. Today, IT has become the nerve
center of a business and has engulfed every
other business process, so much so, without IT,
61
businesses of the day might come to a grinding
halt. Being a powerful engine in driving daily
operations of the organization, IT also helps
in building innovative products and solutions.
IT is being used as a catalyst to drive business
growth.
However, in the recent times a sliding
economy has coerced drastic cuts in IT budgets.
Typically IT is viewed as a cost center and
naturally so, IT budgets were one amongst
the first to take a hit. No new projects got
initiated, new IT procurements were put on
hold and various means of cutting costs were
explored. So did this recession translate to? IT
was used as a tool to beat downtrends. Many
organizations leveraged IT to improve their
effciencies, build new tools, improvised on
their automation capabilities and re-trained
staff on new technologies. This helped them
reduce their operational expenditure. No new
projects translated into the provision of a good
platform for the organization to launch austerity
measures. In all this, SIs were the key players,
who helped their customers by collaborating
and co-opting with them in creating new things.
Many SIs were part of the innovation drive
within organizations, helping customers with
technology support. The multi-faceted and
multi-domain talent of the SIs was brought
to the table along with the business talents of
the organization, to achieve such co-created
products and services.
LAMP, an acronym for Linux, Apache,
MySQL, Perl/PHP/Python, an open source
software stack is a classic example of a co-
created invention. Though it falls under the
category of distributed co-creation by like
minded professionals coming together and
developing a set of software components,
LAMP has been used by many organizations
as a reference architecture stack. There have
been many customizations to LAMP thereby
making it work with other operating systems
as well.
Co-Creation
Continuum
Social
Media,
Web 2.0
Growth
Values Outcomes
Open Innovation
Crowd Sourcing
Social Networking
Customer
Suppliers
Consumers
Common Objectives Business Strategy
the eco-system
Figure 1: Components of Enterprise Architecture Source: Infosys Research
62
Even IT service providers and SIs
themselves were forced to undertake such
i nnovat i on wi t hi n t hei r organi zat i ons.
Compelled to provide solutions to customers
at lower costs and slashed timelines, IT service
providers developed innovative solutions
rangi ng f rom re- usabl e component s t o
productivity improvement tools to co-creation
platforms that helped them provide cost-
effective solutions to their clients. Innovation
was a major focus of such organizations in
the last couple of years. Even newer pricing
models that would provide mutual business
benefits gained momentum in these years.
Internet adoption and a new wave of digital
media have fuelled growth of innovation
further. Digital platforms brought people,
world over, much closer and have helped
the collaboration wave thus resulting in new
innovation models like open sourcing, crowd
sourcing and the like.
Co-created products and services may
be the critical survival tools for now, but in
the long run are sure to be absorbed into the
mainstream operations of organizations; to
bring in the required improvement in effciency;
and increase the brand value of companies. In
fact, organizations have started to come out
of the mindset that innovation is at its best
form during a crisis situation or an economic
meltdown. Innovation has become a routine
in most of the organizations and certain
organizations have also constituted Innovation
Offce to drive such initiatives [2].
With IT being increasingly seen as
the key driver and enabler for the business,
organizations are constantly in the process
of refning their IT footprint to make it more
agile and effcient to support business growth.
There is a drastic shift in the way IT operates in
an organization compared to a few years ago.
Today, organizations have come to realize that
it would be much more benefcial and result
oriented to involve relevant stakeholders to the
exact the best results out of an offering.
ROLE OF SIS IN CO-CREATION
In the true sense, SIs integrate multi-vendor
variants of products and technologies, multiple
and discrete product and technology systems to
bring up an integrated platform using which
business services are delivered. Integration of
such diverse range of products and solutions
requires multiple skillsets and to most extent SIs
are vendor neutral. Apart from the IT domain,
SIs also carry business domain knowledge
and expertise, enabling them to look at the
big picture and advise clients on business-
technology integration. Integration of systems
can occur in any scale and size starting from a
small time core IT system installation to setting
up a large IT application and its subsystems for
a corporate entity.
The role of SIs within a customer
organization is slowly transforming from
that of a normal vendor to a trusted IT
advisor and partner. Successful SIs have
compl et e knowl edge of t he cust omer s
business strategy, their business and IT
processes and their enterprise architecture
landscape. By implementing systems that
are mainly intended to improve efficiencies
and support growing business needs, SIs
acquire in-depth knowledge of the business
domain and are well positioned to partner
with the customer to (a) resolve their existing
issues; and (b) develop mutually beneficial
products and services. Both parties benefit
from such a collaborative effort. For the
customer, co-creating with someone who
knows their business problem is all the more
comforting and working with them becomes
63
easier. Moreover the model provides for
de-risking certain business risks.
The value creation platform has the SIs in
the core and an extended eco-system of partners,
vendors and even society at large. Setting
up a customer co-creation vehicle invariably
involves a pre-established and concretized
relationship between the two entities, with the
SI understanding pain-points and concerns that
customer has. A problem area is defned and
both parties work towards solving the problem
on hand. This cycle could see new products
and services (aimed at end consumers), tools
and uti l i ti es, re-usabl e components and
frameworks (aimed at improving their internal
productivity), co-branding activities (aimed
at increasing the market share and the brand
value) being developed and launched. Volume
or scale is immaterial as long as the business
objectives are met. There are various business
models under which co-creation activities will
be taken up, which decides how the cost and
beneft sharing is to be done. In addition, the
main bone of contention could be the ownership
of the intellectual property right (IPR) of the
resultant co-created product. Hence, suffcient
care and attention has to be paid to this aspect.
Technology infusion is a key area in
which SIs bring in competence and skills. A
classic case of a new technology area where
SIs can potentially co-opt with organizations
i s i n the cl oud computi ng space. Wi th
increasing focus towards cloud enabling the
IT infrastructure, most organizations are
trying to see how cloud can be leveraged
to increase business traction. The concept
of cloud still being in its nascent stage, SIs
can help organizations build and operate a
(private) cloud environment. There are ample
opportunities for creating new solutions
around cloud, thereby realizing potential
benefits like cost reduction, efficiency and
agility amongst other benefits.
CO-CREATION WITH CUSTOMERS:
CHALLENGES
Bei ng a col l aborati ve effort between an
organization, its end users and external vendors
and more importantly, internet and social
media being one of the actors, there are certain
challenges from legal and compliance perspective
that enterprises will have to encounter in a co-
creation effort. Some of the key and critical
challenges are listed below in Figure 2.
CRITICAL SUCCESS FACTORS FOR
CO-CREATION
Value creation through customer co-creation
is based on the premise of collaboration,
interaction and trust. Both parties are in
Figure 2: Challenges Faced During Customer Co-creation
Protecting IP Rights, copyrights
Protecting Brand value
Privacy & Non- disclosures
Funding of initiatives
Contracts & re-imbursement
Revenue and Cost Models
Legal Financials
Defining the Objectives
Managing expectations
Managing Participation
Management
Technology
Source: Infosys Research
64
equal footing in such an agreement and the
success (or even failure) depends on the level
of relationship and interactions between
customer and SI. Some of the key factors that
influence success of such initiatives are listed
below.
Establishing mutual trust and according
respect to other partner(s). It is like a
marriage and the relationship has to be
treated and nurtured properly.
SI s s houl d devel op a deep and
credible understanding of customer
requirements.
Staying focused on goals is a must.
Innovation activity in an organization
might lead to delays, dynamically
changi ng requi rement s and even
shortage of funds and resources. Also
there could be challenges from various
stakeholders and many opportunities for
distraction. Focusing on end objective
therefore is very critical.
Always having the big picture in mind.
Innovation initiatives are initiated with
long term goals in mind. Quick wins
might be impossible.
Involving all relevant stakeholders, be
they employees, vendors or partners.
Their concerns and voices have to be
heard.
Recogni zi ng and r ewar di ng t he
deserving. What is in it for me, is the
key thought of a contributor. Hence
giving something back to them keeps
them motivated.
Value Focus Characterization Example
Monetary
Business
Growth,
Revenue
Business oriented, making money
Transactional metrics, and usually based on short term needs,
but intending to generate sustainable revenue over longer
time horizon
Productized market segment, involves commodity transactions
Life-stage results
Large scale
production of
computers,
incorporating
user reviews and
feedback
Experience
User
Experience,
Customer
Centricity
Based on personal experiences, increased the depth of
customer engagement, derive customer satisfaction and
delight
Relationship oriented, emotional connect at times and more
importantly, a personal touch involved
Knowledge artifacts, results in re-use
Life-time results
Users choosing
their own
personalization
on the computer
confguration
Social
Humanity,
Well-being
of the
society
Humanity,
Well-being
of the
Society
Aimed towards social cause, humane approach, bettering the
way of living
Contribution for a social cause, giving back to the society is
the fundamental approach
Monetary benefts not the end result
Results carried into the next generation
Creating a web-
enabled platform
for a social cause.
For e.g., rural
education
Table 1: Values of Co-creation Source: Infosys Research
65
VALUE CREATION
Any innovation initiative should generate value
for all stakeholders and more importantly,
they should be able to perceive it. But what is
value in the co-creation continuum? It can be
described as a phenomenon which a consumer
experiences and feels that she has got a good
worth for whatever has been spent, in terms
of money, time, etc. Value can be tangible like
monetary value or intangible like experience,
feeling, etc. There are three broad values that
can be defined in the co-creation context as
shown in Table 1. And more so, in a co-creation
environment with multiple stakeholders,
it is highly imperative that the values thus
generated be well articulated.
Co-creation activities involving an
SI and customer leads to any one or a combination
of values and is refected in the output generated
by such activities. More often than not, such an
initiative tends to look at monetary beneft over
a period of time, either tangibly or otherwise.
Initiatives for creating social value are always
associated with a social responsibility and
considered as a means of bettering the human
value of life. Co-creation activities leading
to social value are typically spearheaded by
NGOs with contributions from corporate and
technology firms. While the first two types
intend to capture direct customer involvement,
those striving to create social value involve
multiple actors and bigger challenge is that,
most of times, will not even be aware of what
the end product / service would be.

CONCLUSION
Co-creation has acquired a fresh outlook.
The current wave of innovation is a budding
discipline, collaborative creativity, focusing
on people and aiming to enhance the value
provided to consumers. In a marked difference
from a normal creation, co-creation involves
end users or consumers themselves, who get
an opportunity to create or at least be heard on
their wants in the creating process. Co-creation
puts the responsibility of creating things in the
hands and minds of people who would be using
them. It helps improve user experience, increase
customer satisfaction and build a strong brand
value. Innovation through co-creation is an
important strategy to improve business growth
and customer retention.
SIs help organizations realize the co-
creation dream and rightfully so by bringing
technology and domain skills. SIs help create
and build the necessary technology platform
from where organizations can start rolling out
better services. And more importantly, SIs help
in advising and implementing niche technology
platforms for the end customers. This forms an
important part of realizing the much advocated
business-IT alignment.
REFERENCES
1. Prahalad, C. K. and Ramaswamy, V.
(2000), Co-opting Customer Competence,
In Harvard Business Review. Available
at http://hbr.org/2000/01/co-opting-
customer-competence/ar/1.
2. Davi s, I. (2009), The new normal ,
Mckinsey & company. Available at
https://www.mckinseyquarterly.com/
The_new_normal_2326#.
3. Levn, P. & Hol mstrm J. (2008),
Consumer co-creation and the ecology
of innovation: A living lab approach,
Department of Informati cs, Ume
University, Sweden. Available at http://
umu.diva-portal.org/smash/record.
jsf?pid=diva2:311009.
4. PRI EM, R. ( 2007) , A Cons umer
Perspective on Value Creation, Academy
66
of Management Review, Vol. 32, No. 1,
pp. 219235.Available at http://faculty.
fuqua. duke. edu/~charl esw/s591/
Bocconi-Duke/Bocconi/s6_Demand_
View/priem.pdf.
5. Winther, T., (2002), Value Creation
and Proft Optimization. Available at
http: //www. exinfm. com/pdffiles/
ValueCreation.pdf.
6. Luftman, J. (2003), Measure Your Business
IT Al i gnment, www. opti mi zemag.
com, December 2003. Avai l abl e at
http: //www. bita-center. com/pdf/
Alignment_Maturity_Summary.pdf
67
Index
Action Manager 9
Applied Information Economic, also AEI 24
Analytical Hierarchy Process, also AHP 33
Architecture 6-8, 9-14, 22, 25, 32, 35-37, 39-42, 45-54
Center of Excellence 25
Development Method, also ADM 22
Enterprise, also EA 36
Enterprise Solution, also ESA 53
Governance Process 42
Governance 25
Model Driven, also MDA 53
Performance 32
Solution 54
Service Oriented, also SOA 6-8, 9-14,
35-37, 39-42, 45-52
Systems Integration, also SIA 36
Tradeoff Analysis Method, also ATAM 32
Business 6, 7, 19, 21-27, 35-40, 42-43 50, 53
Activity Monitoring, also BAM 39
IT Alignment, also BITA 21-27
Process Management, also BPM 6, 7,
19, 35, 40, 53
Rules Management 50
Transformation 35-39, 42-43
Value Index, also BVI 23
Co-Creation 59-65
Commercial-Off-The-Shelf, also COTS 5, 46
Complex Event Processing, also CEP 7, 8, 9, 10,
CRM System 17
E-Commerce 29-31
Enterprise 10-13, 17, 31, 46-47, 49-52
Application Integration, also EAI 7, 10
Content Management 17
Service Bus, also ESB 10-13, 31, 46-47,
49-52
ERP System 17
Event 9,12
Driven Architecture, also EDA 12
Pattern Modeler 9
Pattern Recognition Engine 9
Framework 21-27
COBIT 24
COSO 24
EA 25, 27
FEA 21, 22
ITIL 24
PMBOK 26
PRINCE2 26
TOGAF 21-27
ZACHMAN 21-27
Integration Platforms 10
Intelligence 9-10, 17
Artifcial 10
Business 17
Real-Time 9
IPv6 Networks 3
Mainframes 5, 12
Management 13, 15, 17-19, 23-24, 26-27 31, 37,
39-40, 45, 47-49, 51, 54-55
Access 18
Change 19, 54
Confguration 26
Content 31, 40
Customer Service 15
Data 48
Data Warehousing 17
Exception 48
Feedback 37
Innovation 27
IT Service 24
68
Inventory 45
Knowledge 17, 23
Master Data, also MDM 17
Order 31
Organizational 26
People 19, 54
Policy 47, 51
Process 39, 48-49, 54
Program 19, 23
Project 13,26
Risk 55
Telecom Order 39
Time 37
Transaction 49
Service 40, 47-49
User 48
Vendor 26
Message 5, 9
Bus 9
Oriented Middleware 5
Net Present Value, also NPV 24
Proactive Performance Engineering 29-32
Queuing Network Model, also QNM 31
Representative State Transfer, also REST 49
Return on Investment, also ROI 36
Rule Engine 9
Screen Scraping 5
Simple Operation Access Protocol, also SOAP 47
Software Development Life Cycle, also SDLC 30
Standards Information Base, also SIB 26
Strategic Alignment Maturity Assessment
Model, also SAMM 21
System/s 10-11, 38-39, 47
Business Support, also BSS 38
Central Nervous, also CNS 11
Cognitive 10
Enterprise Messaging, EMS 47
Operations Support, also OSS 38
Peripheral Nervous, also PNS 11
Telecom Billing, also TBS 39
Total Economic Impact, also TEI 23
Transaction Wrapping 5
Transformation Program 30, 36-38, 41-43, 55-57
Unstructured Business Processes 9
View 55-57
Business 55
Coexistence 55
Integration 56
Risk 57
SLA 56
Technology 57
Universal Description
Discovery and Integration, also
UDDI 50-51
BUSINESS INNOVATION through TECHNOLOGY
Editorial Office: Infosys Labs Briefings, B-19, Infosys Ltd.
Electronics City, Hosur Road, Bangalore 560100, India
Email: InfosyslabsBriefings@infosys.com http://www.infosys.com/infosyslabsbriefings
Infosys Limited, 2012
Infosys acknowledges the proprietary rights of the trademarks and product names of the other companies
mentioned in this issue. The information provided in this document is intended for the sole use of the recipient
and for educational purposes only. Infosys makes no express or implied warranties relating to the information
contained herein or to any derived results obtained by the recipient from the use of the information in this
document. Infosys further does not guarantee the sequence, timeliness, accuracy or completeness of the
information and will not be liable in any way to the recipient for any delays, inaccuracies, errors in, or omissions
of, any of the information or in the transmission thereof, or for any damages arising therefrom. Opinions and
forecasts constitute our judgment at the time of release and are subject to change without notice. This document
does not contain information provided to us in confdence by our clients.
Editor
Praveen B Malla PhD
Deputy Editor
Yogesh Dandawate
Graphics & Web Editor
Rakesh Subramanian
IP Manager
K V R S Sarma
Marketing Manager
Gayatri Hazarika
Online Marketing
Sanjay Sahay
Production Manager
Sudarshan Kumar V S
Database Manager
Ramesh Ramachandran
Distribution Managers
Santhosh Shenoy
Suresh Kumar V H
How to Reach Us:
Email:
InfosysLabsBriefngs@infosys.com
Phone: +91 40 44290563
Post:
Infosys Labs Briefngs,
B-19, Infosys Ltd.
Electronics City, Hosur Road,
Bangalore 560100, India
Subscription:
infosyslabsbriefngs@infosys.com
Rights, Permission,
Licensing and Reprints:
praveen_malla@infosys.com
Infosys Labs Briefngs is a journal published by Infosys Labs with the
objective of offering fresh perspectives on boardroom business technology.
The publication aims at becoming the most sought after source for thought
leading, strategic and experiential insights on business technology
management.
Infosys Labs is an important part of Infosys commitment to leadership
in innovation using technology. Infosys Labs anticipates and assesses the
evolution of technology and its impact on businesses and enables Infosys
to constantly synthesize what it learns and catalyze technology enabled
business transformation and thus assume leadership in providing best
of breed solutions to clients across the globe. This is achieved through
research supported by state-of-the-art labs and collaboration with industry
leaders.
About Infosys
Many of the worlds most successful organizations rely on Infosys to
deliver measurable business value. Infosys provides business consulting
technology, engineering and outsourcing services to help clients in over
32 countries build tomorrows enterprise.
For more information about Infosys (NASDAQ:INFY), visit www.infosys.com
Infosys Labs Briefings
NOTES
NOTES
NOTES
ANIL PRASAD KURNOOL is a Senior Technology Architect with ECS vertical of
Infosys. He can be reached at AnilPrasad_Kurnool@infosys.com.
ANIRBAN GHOSAL is a Principal Technology Architect with the RCL business unit
of Infosys. He can be contacted at anirbang@infosys.com.
BALAJI SUBBARAMAN was a Technical Architect with Infosys Cloud Computing
Practice.
GANESH JAYARAMAN is Senior Consultant with the Communication domain and
process consulting practice. He can be reached at Ganesh_jayaraman01@infosys.com.
KARTHICRAJA GOPALAKRISHNAN is a Senior Technology Architect with
the Financial Service and Insurance Business Unit of Infosys. He can be contacted at
Karthicraja_G@infosys.com.
LADISLAV GATA was a Senior Technology Architect with the Consulting & Systems
Integration Practice for Retail Business Unit of Infosys.
N. VIJAYKUMAR is a Principal Technology Architect with Infosys Cloud Computing
Practice. He has around 18 years of experience in the IT Infrastructure domain, both in
consulting and implementation.
NARAYANAN CHATHANUR is a Senior Technology Architect with the FSI business
Infosys. He can be contacted at narayanan_cs@infosys.com.
PRAKASH RAJBHOJ is a Principal Technology Architect with the rcl business unit of
infosys. He can be reached at prakashr@infosys.com.
RAMKUMAR DARGHA is a Principal Technology Architect with Cloud Computing
practice. He can be contacted at Ramkumar_Dargha@infosys.com.
ROHIT CHRISTOPHER Rohit Christopher is a Senior Technology Architect with Cloud
Computing practice. He can be contacted at rohit_christopher@infosys.com.
SAURABH JOHRI a Senior Manager in the Corporate Planning unit of Infosys.
SAURABH KUMAR MISHRA is a Technology Architect with Infosys cloud
computing practice. He is currently focusing on Big Data Solutions and can be contacted at
saurabhkumar_mishra@infosys.com.
SENTHIL KUMAR PR is a Senior Technology Architect with the Manufacturing
vertical of Infosys. He can be reached at Senthilkumar_pr@infosys.com.
SUCHIT BHATWADEKAR is a Principal Technology Architect with thethe
Consulting and Systems Integration Practice of Finacle at Infosys. He can be contacted at
suchit_bhatwadekar@infosys.com.
The earlier half of IT evolution was primarily focused on automating
time consuming and labor intensive tasks. Later, systems were built to
automate all and sundry organizational processes. as information systems
made workforce productive, many state-of-the-art systems were built in
silos targeting a particular problem. These systems were built using varied
languages and on varied environments. a need was then felt to integrate
these disparate systems to build a system that has best of all worlds
functionality. Systems Integration (SI) therefore emerged as a field that
brought together disparate systems to operate as one single system.
Today the role of systems integrators is becoming increasingly important
as more and more systems are being developed and such newly built
systems are to be in sync with the existing systems. The job of the systems
integrators is therefore that of developing interfaces to new and existing
systems such that they glue together as one system.
modernization of existing platforms has in a way become the need of the
hour and no sane enterprise can go about engaging in a modernization
initiative minus a clearly defined methodology. One cannot do away with
existing legacy systems, not only because of the huge replacement costs that
enterprises will have to incur, but also because of the inherent advantages
that they have in terms of being storehouses of enterprise processes and
data. Therefore, a better and cleverer way of embracing modernity is by
integrating legacy systems with todays enterprise applications. Enterprises
now are therefore not to shed their legacy systems but modernize them
leveraging SI expertise.
SI has traditionally been looked upon as a deployment and maintenance
function. However, todays SI has the capability to partner enterprises in
their business transformation journey. In this collection we have papers that
shed light on how systems integrators can play a key role in modernizing
legacy and helping transform organizations into smarter organizations.
your feedback is important to us. as usual, do write to us with your
suggestions on how we can further improve your reading experience.
yogesh Dandawate
Deputy Editor
yogesh_dandawate@infosys.com
Authors featured in this issue
Infosys Labs Briefings
Advisory Board
Anindya Sircar PhD
Associate Vice President &
Head - IP Cell
Gaurav Rastogi
Vice President,
Head - Learning Services
Kochikar V P PhD
Associate Vice President,
Education & Research Unit
Raj Joshi
Managing Director,
Infosys Consulting Inc.
Ranganath M
Vice President &
Chief Risk Officer
Simon Towers PhD
Associate Vice President and
Head - Center for Innovation for
Tommorows Enterprise,
Infosys Labs
Subu Goparaju
Senior Vice President &
Head - Infosys Labs
modernize Legacy with
Systems Integrators
For information on obtaining additional copies, reprinting or translating articles, and all other correspondence,
please contact:
Email: InfosyslabsBriefngs@infosys.com
Infosys Limited, 2012
Infosys acknowledges the proprietary rights of the trademarks and product names of the other
companies mentioned in this issue of Infosys Labs Briefngs. The information provided in this
document is intended for the sole use of the recipient and for educational purposes only. Infosys
makes no express or implied warranties relating to the information contained in this document or to
any derived results obtained by the recipient from the use of the information in the document. Infosys
further does not guarantee the sequence, timeliness, accuracy or completeness of the information and
will not be liable in any way to the recipient for any delays, inaccuracies, errors in, or omissions of,
any of the information or in the transmission thereof, or for any damages arising there from. Opinions
and forecasts constitute our judgment at the time of release and are subject to change without notice.
This document does not contain information provided to us in confdence by our clients.
VOL 10 NO 2
2012
SySTEmS
INTEgraTION
I
n
f
o
s
y
s

L
a
b
s

B
r
i
e
f
i
n
g
s
S
Y
S
T
E
M
S

I
N
T
E
G
R
A
T
I
O
N
V
O
L

1
0

N
O

2

2
0
1
2
Subu Goparaju
Senior Vice President
and Head of Infosys Labs
At Infosys Labs, we constantly look for opportunities to leverage
technology while creating and implementing innovative business
solutions for our clients. As part of this quest, we develop engineer-
ing methodologies that help Infosys implement these solutions right,
frst time and every time.
Infosys Labs Briefings